From ltiu at alumni.sfu.ca  Wed May  1 01:50:21 2002
From: ltiu at alumni.sfu.ca (ltiu)
Date: Tue, 30 Apr 2002 22:50:21 -0700
Subject: Summary: sharing tape drive across network.
Message-ID: <20020501054857.FDNA1332.priv-edtnes11-hme0.telusplanet.net@there>

rsh has to be running. 

This is set in inetd.conf and you need a /.rhost file with the line

192.168.1.2 root

Thanks.

ltiu


Subject: sharing tape drive across network.
Date: Tue, 30 Apr 2002 22:26:40 -0700
From: ltiu <ltiu at alumni.sfu.ca>
To: sunmanagers at sunmanagers.org

Hello,

The Sun SA-238 manual mentions that you can do a remote ufsdump using:

ufsdump  0uf  192.168.1.1:/dev/rmt/0  /etc

What it doesn't mention is how do you allow computer 192.168.1.2 to use
computer 192.168.1.1's tape drive over the network?

Could someone help me here?

Thanks.

ltiu
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers

-------------------------------------------------------
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From raji2 at hotmail.com  Wed May  1 08:03:00 2002
From: raji2 at hotmail.com (raji kuruganti)
Date: Wed, 01 May 2002 12:03:00 +0000
Subject: Summary:umount not unmounting the file system
Message-ID: <F42d4qVEs7cEsq7Ogve00007cb6@hotmail.com>

All,

I got to reboot the box to fix it. many replied if there are any mount 
issues or automount etc., but there is no auto mount running either.

thanks to somesh for pointing me to the bug in fuser i found few open files 
as stated in the bug id 4007329 on sunsolve.sun.com. but could not solve the 
issue except by a reboot..

thanks all once again

lakshmi
********************************************************************
My Original post

all,

I need help in unmounting file system.
am  trying to unmount a file system (used for some application) on a sun
box
running 2.6 which is always reporting that the file system is busy .
the file system is not shared, not a home directory for any user and
fuser
doesn't report any users on it.and "lsof" doesn't report any open
files...any help is highly appreciated.

thanks
lakshmi
_______________________________________________

_________________________________________________________________
Join the worlds largest e-mail service with MSN Hotmail. 
http://www.hotmail.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From mgreene at aci.on.ca  Wed May  1 08:51:16 2002
From: mgreene at aci.on.ca (Marco Greene)
Date: Wed, 1 May 2002 08:51:16 -0400
Subject: SUMMARY: Digi PortServer II Rack
Message-ID: <000001c1f10e$e2606990$ccc8a8c0@pepstep>

Only really got once response....

Here it is....I have decided to go ahead and order it...I will let you
know if anything doesn't work as advertised.

> 
> I am looking at purchasing one of these puppies to handle my Sun
> servers.  One thing that concerns me a bit though is the fact 
> that they
> are not too clear on which cable I need.
> 
> I have a mixture of E4500's, E3000, E420R, E250s.
> 
> So this posting is really for three questions:
> 
> 1)  Has anyone had any good (or bad) experiences with this terminal
> server.
I have one working fine here. I have no issues with it.
> 
> 2)  Which cable do I need to have made up as I need longer cables then
> Digi provides.
I'm using custom made 8 pin cables ( with altpin enabled in the
configuration) 
for DB25 machine (most sun servers)
http://support.digi.com/support/techsupport/common/cables/async/PC_term_
rj45
-8pin-altpin--db25f.htm 
for Netra:
http://support.digi.com/support/techsupport/common/cables/async/rj45-8pi
n-al
tpin--cisco-rj45-8pin.htm

> 
> 3)  I used to work with the Annex terminal server and for 
> that I didn't
> have to do any special setup at the server end.  Just configure the
> ports on the Annex and plug it into the serial port on the server.
> Since there was no monitor/keyboard attached, the server defaulted to
> TTYa...and all I had to do was telnet to a specific port and 
> voila....a
be aware that  PS II doesn't have port buffering like annex have ( to
show
you a couple of line before you connected) 
the portserver TS with latest firmware should have that ( as promised by
digi) + SSH server
Digi says they have no plan to add that features to ths PS II

> console.  What is this realport driver all about with the 
> digi product?
never had to use it. 
> I would assume they work pretty close to the same.



> _______________________________________________
> sunmanagers mailing list
> sunmanagers at sunmanagers.org
> http://www.sunmanagers.org/mailman/listinfo/sunmanagers
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From Robert.Boldin at bellsouth.com  Wed May  1 09:13:15 2002
From: Robert.Boldin at bellsouth.com (Boldin, Bob)
Date: Wed, 1 May 2002 09:13:15 -0400
Subject: Summary: Fully Documented Full System Restore
Message-ID: <7716ECBE0D9ED411B18300508BE362DA03434F4A@BLSMSGPRV11>

Managers,

	I received 2 replies to the initial Inquiry I sent out the other week - Thanks Tony and David.  One was a 50 page document that was developed and covered the restore of systems in exquisite detail.  It did not however address the use of Netbackup.  Our internal sa has also sent me his very rough draft of the procedures he went through to get our particular system up which is another 20+ pages.  As time allows and I receive any other input I will attempt to assemble these pieces into a final document that I will make available to anyone having an interest.  If there is overwhelming interest will see if I can post it somewhere and then publish it's link to the group.

As always thanks for the help!

>  -----Original Message-----
> From: 	Boldin, Bob  
> Sent:	Friday, April 19, 2002 8:23 AM
> To:	sunmanagers at sunmanagers.org
> Subject:	Fully Documented Full System Restore
> 
> Managers,
> 
> 	Was wondering if anyone has a complete step by step detailed procedure to perform a full system restore.  We have gotten the system mostly back at this time but have felt ALOT of pain along the way for not having a procedure.  Fortunately this happened on the test environment so no outage was incurred and we had time to figure it out.  We will also be developing a document on this procedure and I will post it but figured I'd ask in the mean time to see what you all had.  Of course I will post the procedure as a Summary.
> 
> Veritas encapsulated root disk (vxinstall type that put everything into rootdg including /opt)
> Veritas Netbackup was backup method
> System Running Sun Cluster 2.2 / Oracle DB
> Shared D1000 (Nvramc used to set scsi-initiator) ** Story in itself {use of set-defaults can really hurt in this case}
> Jumpstart or boot server not available - only a cdrom
> 
> Some things we have learned.
> 1.) You definately can not restore over top of an active OS.  Makes for cpu panics
> 2.) The Use of Netbackup makes it interesting as you do not have the ability to restore while booted from cdrom
> 3.) The inclusion of everything in rootdg is not a good thing (Have now bought the Sun blueprints book "Boot Disk Management" and learned alot - sure wish whoever built these systems had read it also)
> 4.) You have to set specific options for the Netbackup Restore so that Links work
> 5.) You have to un-encapsulate Veritas
> 6.) It's not easy to try and fix a broken layout on top of putting the system back into operation
> 
> Thanks and appreciation for any insights,
> Bob Boldin
> 
> 
**************************************************************************************************************************************************
"The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential, proprietary, and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from all computers."
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From mbox-l at cleopas.stat.vt.edu  Wed May  1 10:04:29 2002
From: mbox-l at cleopas.stat.vt.edu (Mike Box)
Date: Wed, 1 May 2002 10:04:29 -0400 (EDT)
Subject: SUMMARY:  Solaris 8 - UFS logging or logiing via DiskSuite
  Trans Metadevices
Message-ID: <200205011404.KAA11103@cleopas.stat.vt.edu>

I share with you the response to questions that I asked Sun tech support
regarding filesystem logging.


Mike Box                       Phone: (540)231-9506
Systems Administrator            Fax: (540)231-3863
Department of Statistics      E-mail: Mike.Box at vt.edu
Virginia Tech

-----------------------------------------------------------------------------

>Can I implement UFS "logging" (via /etc/vfstab) on RAID5 and mirrored 
>filesystems without creating Trans Metadevices as described in 
>"Solstice DiskSuite 4.2.1 User's Guide" pages 91-99?  
>What are the pros and cons?

Yes you can enable UFS logging on raid5 and mirrored filesystem 
via solaris without creating transmetadevices.


UFS logging and Disksuite's trans metadevice logging serve the same
purpose which is
to store the UFS transactions in a log before the
transactions are applied to the file system.  However the two have
some differences as listed below:

a. The Disksuite logging is available with the Solstice Disksuite
   software while the UFS logging is part of the base OS from 
   Solaris 7.

b. The logging device to which the UFS transactions are initially
   stored is a separate metadevice object in Disksuite.  The log is
   allocated from free blocks on the same file system with UFS
   logging.

c. Since the logging device is a separate device with Disksuite, its
   size can be changed after configuration and also can be mirrored
   for additional reliability.  This is not possible with UFS logging
   as the logging device is part of the UFS filesystem.

d. A single logging device can be shared between trans metadevices
   with Disksuite.  The log is allocated for every individual UFS
   filesystem in case of UFS logging.

e. The UFS transactions can be flushed out of the log and
   written to the master file system using -f option of "lockfs" 
   command with UFSlogging.  This option is not available with 
   disksuite logging.



>Also, I mirrored / (root) partition which already had logging active.
>(Oversight) It seems to be working okay - no messages complaining otherwise.  
>Is this okay or have I "loaded a bullet in the chamber" of which I will
>become aware later at an inopportune time?


You can enable UFS logging on root NOT disksuite logging ( trans metadevices)
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From donunix at yahoo.com  Wed May  1 12:43:49 2002
From: donunix at yahoo.com (Don Jones)
Date: Wed, 1 May 2002 09:43:49 -0700 (PDT)
Subject: SUMMARY: rsh and awk
Message-ID: <20020501164349.60033.qmail@web21301.mail.yahoo.com>

Thank you all:
Answers:

rsh cobra "ps -ef | grep bb | awk '{print \$2}'"
or 
rsh  cobra "(ps -ef | grep bb | awk \'{print $2}\')"
or
rsh -n cobra "ps -ef | grep bb | awk '{print $2}'"
or simply
rsh cobra ps -ef | grep bb | awk '{print $2}' 
with no quotes at all..

-DJ


------------------------------------------
On Mon, 29 Apr 2002 16:05:31 -0700 (PDT),
donunix at yahoo.com wrote:
> Hello Everyone,> > I have problem with awk working
with rsh:> 
> rsh cobra "ps -ef | grep bb | awk '{print $2}'"> >
Only this execute:> 
> csh -c ps -ef | grep bb | awk '{print }'> 
> Why $2 does not execute?! how can I fix it?
Yahoo! Health - your guide to health and wellness
http://health.yahoo.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From scotrn at cox.net  Wed May  1 15:42:35 2002
From: scotrn at cox.net (scotrn)
Date: 01 May 2002 15:42:35 -0400
Subject: SUMMARY: Perl Kstat: Module won't compile.
In-Reply-To: <OFEDC51540.C17A2E28-ON86256BAB.006A1378@uscmail.dcn>
References: <OFEDC51540.C17A2E28-ON86256BAB.006A1378@uscmail.dcn>
Message-ID: <1020282155.6451.1.camel@sywn001.engrs.infi.net>

This problem was fixed on all systems by downgrading to 
Perl  5.005_03. The problem seems to be focused around 
changes to the perl header file XSUB.h in 5.6 and up 

http://www.perldoc.com/perl5.6/pod/perlxstut.html



hing to try is setting POLLUTE=1 on the:
> perl Makefile.PL 
> step.  A shot in the dark but it worked for me on a similar perl error. 
> Hmm, and I see it in my Solaris-0.5 Makefile
> 
> a
> 
> Andy Bach, Sys. Mangler
> Internet: andy_bach at wiwb.uscourts.gov 
> VOICE: (608) 261-5738  FAX 264-5030
> 
> It is when I struggle to be brief that I become obscure. (Horace)
>                      [ or, in perl: -- or & ]
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From mangesh.v at xaalliance.com  Wed May  1 22:05:10 2002
From: mangesh.v at xaalliance.com (Mangesh)
Date: Thu, 2 May 2002 10:05:10 +0800
Subject: Summary : Filesystems
Message-ID: <003701c1f17d$ca2a9980$662aa8c0@Mangesh>

Hi All,

       Thanks to Tony Walsh, GreenLand Simon, Andrew Stueve for there quick
replies & help..........
       Below given are the replies.............

My query was ...

----- Original Message -----
From: "Mangesh" <mangesh.v at xaalliance.com>
To: "sunmanager" <sunmanagers at sunmanagers.org>
Sent: Tuesday, April 30, 2002 11:37 AM
Subject: Filesystems


> Hi All,
>
>  I have a E-450 running Solaris 8. It is connected to the two T3 Sun
StorEdge
> ( T3-1 & T3-2 ). RAID 5 has been configured from the  hardware side i.e
T3.
> Both the T3 boxes are mirrored across each other with dual path to the
> server.
>  Now I have 4 filesystems which are layed on the T3-1, which have 7 disks
of
> 73 GB each. Each box having 2 controllers (dual path).
>  I have Veritas Volume manager 3.1.1 & Veritas Cluster 1.3.0.
>
>   I can see the T3-1 box through one controller, as one full chunk of disk
> space ( 7 X 73GB).
>   My vendor wants, that all the filesystems should be mapped to each
physical
> disk of the 7 disks.
>   for eg. if I have 5 filesystems, every filesystem should map to the
> different disk on the T3-1 ( 5 x 73GB ) i.e 5 filesystems, each of 73Gb
each.
>  Is it possible to create 5 filesystems on 5 different physical disk from
the
> T3-1.
>
> Any help or suggestions would be highly appreciated.
> Thanks in advance.
>
>
> Thanks & Regards
> Mangesh
> _______________________________________________
> sunmanagers mailing list
> sunmanagers at sunmanagers.org
> http://www.sunmanagers.org/mailman/listinfo/sunmanagers

Mangesh

The only way you can achieve dual pathing in this setup is to configure the
T3's
as a Partner Pair which you have not mentioned. This means that the 2 bricks
are
connected to each other via Copper Fibre cables across the back of the
arrays.

The T3 will NOT support more than 2 LUNs per tray (or brick) so you are not
able
to map individual disks to an individual LUN as you appear to want to do. If
you
want that kind of limitation then you need to have bought a D1000 (or
possibly an
A1000) so that individual drives are visible to the OS. A T3 has hardware
RAID
capabilities and therefore only presents the LUN to the OS (not drives) and
as
there is a logical limit of 2 LUNs per brick, you can only get 4 LUNs out of
this
configuration.

As there must be 9 drives of the same size in each T3 for a total of 18
drives,
what relevance do the 7 drives have to the 4 file systems you have?

In summary, it is NOT possible to create 5 physically separate filesystems
on 5
physically separate disk drives on any T3. Your best options are to define 1
or 2
LUNs per array in a Partner Pair configuration and then use VxVM to slice
them up
the way you want.

Regards Tony Walsh.

I hope this helps.


Mangesh,

I'm a little confused here: Setting up RAID 5 on each T3 and then RAID 1
between the two sounds a bit excessive, especially when each T3 is highly
redundant (dual power supplies etc). You seem to be using 14 disks to
protect the data held by 6 disks (an 8 disk overhead).

Personally I see RAID 5 as a poorman's mirror: It protects only against a
single disk failure, but is cheaper than full mirroring because less disks
are needed.

You already have plenty of disks, so I'd remove RAID 5 and simply mirror
between the two T3s. This will be more efficient both in terms of available
disk space and in terms of read/writes. It also maintains the ability to
keep runnning if you loose 1 of your T3s, or a combination of disks from
each (as long one disk from each mirror is maintained).

Hope this helps,

Simon


Your 'vendor' should go jump in a lake.  Your 'vendor' shouldn't get
involved in the filesystem locations.  Your 'vendor' is an idiot.

If you have mirrored T3 arrays, then your disk performance should be
outstanding.  Any write contention that the idiot is worried about is a
non-issue.  T3s have hardware the accelorates RAID5 to the point that
it's performance is as good as RAID0.  Mirroring the two T3s does take a
small performance hit on writes, but you more than make up for it in
reads.

The way you have things setup is great.  Tell your idiot vendor to do
his job and stop worrying about yours.  You set it up good.

--
----------------------------------------------------
Andrew Stueve              |   Office 703-886-2606
Team Lead/Sr. Engineer     |
Worldcom                   |   Pager 1-888-454-7594
----------------------------------------------------

Thanks again

Regards
Mangesh.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From mangesh.v at xaalliance.com  Wed May  1 22:26:08 2002
From: mangesh.v at xaalliance.com (Mangesh)
Date: Thu, 2 May 2002 10:26:08 +0800
Subject: Summary : CPU
Message-ID: <007e01c1f180$b8150de0$662aa8c0@Mangesh>

Hi All,

Thanks to Pierre Zimmermann, Santos Ramiro, Greenland Simon, Steve Mickeler,
Jed Dobson, Justin Shaffer, Kevin Buterbaugh,  Adam, Christine, Mike, Andrew
Stueve, Patricio Mora..........
The more help came from Greenland Simon for his detailed summary....
Thanks again everybody.

Everybody asked me to use pbind, psradm, psrset to assign all the processes
you need to the particular CPU's, but may hamper your system if the no of
CPU's is less. You can try it on high end servers like E-6500, SunFire & so
on where you have got many CPU's in hand to play with.

Thanks again...

Regards
Mangesh
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From ltiu at alumni.sfu.ca  Wed May  1 22:43:01 2002
From: ltiu at alumni.sfu.ca (ltiu)
Date: Wed, 1 May 2002 19:43:01 -0700
Subject: Summary: Memory fault.
Message-ID: <20020502024122.QZSI1515.priv-edtnes27.telusplanet.net@there>

Moving the memory sticks around different slots and removing a few sticks at 
a time while leaving some behind solved my problem. One of the memory sticks 
is faulty.

Thanks.

ltiu

----------  Forwarded Message  ----------

Subject: Memory fault.
Date: Wed, 1 May 2002 18:08:14 -0700
From: ltiu <ltiu at alumni.sfu.ca>
To: sunmanagers at sunmanagers.org

Hello guys,

I recently installed 64MB more memory into a SparcStation 5 running Sol5.6.
Total RAM 128MB.

Now it boots with the error:

----------------------------------------------
panic: asynchronous memory fault: MFSR=81802010 MFAR=bba1264
syncing file systems...panic: asynchronous memory fault: MFSR=81802820
MFAR=bc7608f
 2213 static and sysmap kernel pages
panic: asynchronous memory fault: MFSR=81802820 MFAR=bd62360
rebooting...
------------------------------------------------

I went into the OK prompt and typed "test-all".
The memory tested fine. No error messages.

Can anybody give me any clue?

Thanks.

ltiu
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers

-------------------------------------------------------
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From mangesh.v at xaalliance.com  Wed May  1 23:15:27 2002
From: mangesh.v at xaalliance.com (Mangesh)
Date: Thu, 2 May 2002 11:15:27 +0800
Subject: Summary : CPU
References: <006201c1f183$b9340b10$ad01a8c0@bonanza>
Message-ID: <00dc01c1f187$9bc013e0$662aa8c0@Mangesh>

Hi All,

     Sorry for sending the mail again. Many of them have requested me, to
post the orginal messages sent by the people
 whose names were given in my earlier mail.
 Please find the messages below : -

Regards
Mangesh

> -----Original Message-----
> From: sunmanagers-admin at sunmanagers.org
> [mailto:sunmanagers-admin at sunmanagers.org]On Behalf Of Mangesh
> Sent: Wednesday, May 01, 2002 7:26 PM
> To: sunmanager
> Subject: Summary : CPU
>
>
> Hi All,
>
> Thanks to Pierre Zimmermann, Santos Ramiro, Greenland Simon, Steve
Mickeler,
> Jed Dobson, Justin Shaffer, Kevin Buterbaugh,  Adam, Christine, Mike,
Andrew
> Stueve, Patricio Mora..........
> The more help came from Greenland Simon for his detailed summary....
> Thanks again everybody.
>
> Everybody asked me to use pbind, psradm, psrset to assign all the
processes
> you need to the particular CPU's, but may hamper your system if the no of
> CPU's is less. You can try it on high end servers like E-6500, SunFire &
so
> on where you have got many CPU's in hand to play with.
>
> Thanks again...
>
> Regards
> Mangesh
> _______________________________________________
> sunmanagers mailing list
> sunmanagers at sunmanagers.org

----- Original Message -----
From: "Pierre Zimmermann" <Pierre.Zimmermann at tecnomen.fi>
To: "'Mangesh'" <mangesh.v at xaalliance.com>
Sent: Tuesday, April 30, 2002 6:57 PM
Subject: RE: CPU


> Hi Mangesh
> have a look in the man pages of:
> pbind, psradm, psrset and psrinfo
>
> cu
> Pierre


----- Original Message -----
From: "Santos, Ramiro" <Ramiro.Santos at commerzbankib.com>
To: "'Mangesh'" <mangesh.v at xaalliance.com>
Sent: Tuesday, April 30, 2002 7:01 PM
Subject: RE: CPU


> Yes you can,
> in Solaris is possible to bind processes to CPUs.
> see "man pbind"
>
> hoppe this helps
>
>
>
> Yours sincerely,
>
> Ramiro Santos
>
> ZIT P 7.41 Unix Server Implementation
>
> _______________________________________________________
>
> Contact:   Phone:     +49-(0)-69-136-43523
>            Fax:       +49-(0)-69-136-47040
>            E-Mail:    Ramiro.Santos at CommerzbankIB.com
>
> Address:   Commerzbank AG
>            Zentraler Servicebereich IT Production
>            C/S Services Investment Banking - ZIT P 7.41
>            Mainzer Landstr. 151 - DLZ 1, 06.52.0560
>            D-60321 Frankfurt / Main
>


----- Original Message -----
From: "Greenland, Simon D SITI-ITDSEP3" <Simon.D.Greenland at is.shell.com>
To: "'Mangesh'" <mangesh.v at xaalliance.com>
Sent: Tuesday, April 30, 2002 7:09 PM
Subject: RE: CPU


> Mangesh,
>
> Something useful I found on groups.google.com. You will need to start your
> instances from two different scripts (one processor bound, the other not).
>
> I assume you have tried tuning the instance that is causing problems.
>
> Cheers, Simon
>
>
> Two different methods can solve this problem.  First you need to know
> what CPUs are installed (if you had, say an E6500, then processors are
> not necessarily numbered sequentially and you MUST know the actual CPU
> numbers) - type 'psrinfo' and it'll return a list.  Since you said you
> have an E450 and E450's can only have 4 CPU (and you have all 4
> installed) then your CPU numbers must be 0, 1, 2, & 3.
>
> Knowing this you can just type "pbind <cpu#> <process_id>".
>  This will restrict that PID to that CPU.  "pbind" is inherited which
> means that if you pbind a process and that process spawns children then
> all the children will also be bound to that same CPU.  That means that
> if you have a start script, and in the start script you inoke the pbind
> command to rebind the script's own PID (pbind 3 $$ would result in the
> PID being bound to cpu #3 (the fourth CPU)) to a CPU then all the
> processes that the script starts will stay on that same CPU.
>
> This will restrict everything to just one CPU and you wanted them to run
> on two.  You could pbind some processes to, say, CPU 2 and others to
> CPU3, or you could let them float and Solaris would schedule them as
> needed (which would probably yeild the best performance).  To do this
> you psrset.
>
> First create the processor set:
>     psrset -c 2 3            (put processors 2 & 3 in a processor set
> which is created - the command returns the processor set ID).
>
> Then use:
>     psrset -b <processor_set_id> <pid>
>
> If you do have a start script to fire up Oracle and program this into
> the top of the script then every Oracle process started will will be
> bound to the processor set (of 2 cpus) so that they'll float between
> them as Solaris deems best.
>


----- Original Message -----
From: "Steve Mickeler" <steve at neptune.ca>
To: "Mangesh" <mangesh.v at xaalliance.com>
Sent: Tuesday, April 30, 2002 7:16 PM
Subject: Re: CPU


>
> man pbind
>
>

----- Original Message -----
From: "Jed Dobson" <jed at wgtech.com>
To: "Mangesh" <mangesh.v at xaalliance.com>
Sent: Tuesday, April 30, 2002 8:15 PM
Subject: Re: CPU


>
> SRM and/or processor sets.
>
> -jed
>


----- Original Message -----
From: "Shaffer, Justin" <Justin.Shaffer at mlb.com>
To: "'Mangesh'" <mangesh.v at xaalliance.com>
Sent: Tuesday, April 30, 2002 8:34 PM
Subject: RE: CPU


> Depending on the application, and how much context switching you are
> actually doing, this can potentially hurt your performance. You may wish
to
> speak with your database vendor before you play with pbind, if this is a
> production database.
>
> Regards,
>
> Justin
>
> ---
> Justin Shaffer
> Director of Operations
> MLB Advanced Media LP
> 75 Ninth Avenue
> New York, NY 10011
> Office: (212) 485-3499
> Fax: (212) 485-3456
> Cell: (917) 519-5523
> Email: justin.shaffer at mlb.com
>


----- Original Message -----
From: "Kevin Buterbaugh" <Kevin.Buterbaugh at lifeway.com>
To: "Mangesh" <mangesh.v at xaalliance.com>
Sent: Tuesday, April 30, 2002 9:18 PM
Subject: Re: CPU


>
>
> Mangesh,
>
>      Check into processor sets ("man psrset" for starters).   You might
> also want to look at processor binding ("man pbind").  HTH...
>
> Kevin Buterbaugh
> LifeWay
>
> "Anyone can build a fast CPU.  The trick is to build a fast system." -
> Seymour Cray
>


----- Original Message -----
From: "Adam and Christine Levin" <levins at westnet.com>
To: "Mangesh" <mangesh.v at xaalliance.com>
Sent: Tuesday, April 30, 2002 11:00 PM
Subject: Re: CPU


> On Tue, 30 Apr 2002, Mangesh wrote:
> >  I have E-450 running Solaris 8. On the server we have 4 database
instances.
>
> That's a lot of databases on one machine.  The E450 isn't a huge machine
> by any standard.
>
> > I have one query regarding CPU sharing. If one of the database is going
to
> > eat,  lot of system resources & would always have a read/write on it. So
can
> > we allocate no of  CPU's to one database instance & the rest CPU's to
other
> > instance.
>
> You want the psrinfo, psradm and psrset commands.
>


----- Original Message -----
From: "Mike's List" <mikelist at sky.net>
To: "Mangesh" <mangesh.v at xaalliance.com>
Sent: Tuesday, April 30, 2002 11:27 PM
Subject: Re: CPU


> I'm not 100% certain, but what you wanted below is called partition and
> it's only available on Sun's high-end servers, E10K, E12K, E15K, etc.
> where you can assign/partition certain numbers of CPU for certain task.
>
>
> - Mike
>


----- Original Message -----
From: "Andrew Stueve" <andrew.stueve at wcom.com>
To: "Mangesh" <mangesh.v at xaalliance.com>
Sent: Wednesday, May 01, 2002 12:43 AM
Subject: RE: CPU


> This is not a mainframe.  This is a unix box that will allocate
> resources on demand.  You don't allocate CPU to databases, unless you
> are limiting users via SRM.
>
> First, the database can be high load, but unless it is specifically
> written to split all tasks up on separate CPU, then allocating 2 to a
> single process CPU is a waste.  If the database takes over a CPU, then
> it will not try and take a second.  It will just monopolize access to
> the single CPU.  We have 18 databases on a 4 way server, some are high
> load, some are minor.  They take the resources they need as they need
> it. (CPU resources that is, they still have a large SGA in memory.)
>
> If you want to use SRM, then limit the low load databases to 1/2 a cpu,
> and then the high load database will always have a full CPU available
> for it's own use, and can ask for more if it needs to.
>
> --
> ----------------------------------------------------
> Andrew Stueve              |   Office 703-886-2606
> Team Lead/Sr. Engineer     |
> Worldcom                   |   Pager 1-888-454-7594


----- Original Message -----
From: "Patricio Mora" <pmora at cgob.junta-andalucia.es>
To: "Mangesh" <mangesh.v at xaalliance.com>
Sent: Tuesday, April 30, 2002 9:05 PM
Subject: Re: CPU


> I run a 5 min cron job to attach every oracleSID (SID=parasitic database)
process to one of my 18 CPUs, so production will take 17, more or less.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From ahoesch at smartsoft.de  Thu May  2 06:32:42 2002
From: ahoesch at smartsoft.de (Andreas Hoeschler)
Date: Thu, 2 May 2002 12:32:42 +0200
Subject: Summary: Compiling a very simply c programm using a library
In-Reply-To: <8E88578A-5DA1-11D6-9C94-003065CCA582@smartsoft.de>
Message-ID: <EF0000DE-5DB7-11D6-9C94-003065CCA582@smartsoft.de>

Hi all,

thanks to all that replied. The solution was to add

  -lnsl -lsocket

to the compile command.

Regards,

    Andreas
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From Murtuza_Bekhushi at infy.com  Thu May  2 06:20:58 2002
From: Murtuza_Bekhushi at infy.com (Murtuza Bekhushi)
Date: Thu, 2 May 2002 15:50:58 +0530
Subject: SUMMARY:  Freeing a Port on Solaris.
Message-ID: <F0CB3C9983B77E4AB4ADEFA63DAB109F2EE54B@twrmsg03.ad.infosys.com>

	Hi,
		lsof is the solution to this.  I downloaded lsof from
www.sunfreeware.com,  did the make of lsof and ran the following
command,

lsof -i:9030

	This give me the PID of the process which was holding the port.

Thanks to Following people for getting me the way to solution :

Stuart Whitby, Rahul Parasnis, Dan Lowe, Osama Ahmed, Joe Matusiewicz,
Benjamin Ritcey, Dana, Andrew Rotramel, Martynas Buozis, Ramiro Santos,
Douglas Palmer, John England, Hendrik Visage,  Ed Rolison.

		-----Original Message-----
		From: Murtuza Bekhushi
		Sent: Tuesday, April 30, 2002 7:17 PM
		To: sunmanagers at sunmanagers.org
		Subject: Freeing a Port on Solaris.

		Hi all,
			I had some service brought up on a particular
port.   Now the process which brought up the services has gone defunct.
I am unable to find process id which has still held this port.  The
output of "netstat -a |grep 9030" is

		*.9030               *.*                0      0     0
0 LISTEN

			How can I make this port free ?
			I am using "SunOS 5.6 Generic_105181-23 sun4u
sparc SUNW,Ultra-80"

			Pls. help.

		Regards,
		Murtuza.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From Gerard.Henry at cmi.univ-mrs.fr  Thu May  2 09:54:23 2002
From: Gerard.Henry at cmi.univ-mrs.fr (Gerard Henry)
Date: Thu, 02 May 2002 15:54:23 +0200
Subject: [SUMMARY]: adaptec scsi 2940 on solaris sparc
References: <E023B997F74B1E46BB8367A9C548BCFC021B624A@aunty.tecnomen.fi>
Message-ID: <3CD1450F.1000804@cmi.univ-mrs.fr>

Thanks to all who respond:
Pierre.Zimmermann at tecnomen.fi
mikelist at sky.net
krister.bergman at sun.com
Steve Mickeler <steve at neptune.ca>


adaptec don't support this card on sparc
perhaps on sparc/solaris until solaris 2.6 (i have solaris 8) with 
driver adp?


I forgot to consult sunmanagers archives, this question already appears 
and it is the same answer now.


>>-----Original Message-----
>>From: Gerard Henry [mailto:Gerard.Henry at cmi.univ-mrs.fr]
>>Sent: 02 May 2002 15:00
>>To: Sun Managers (E-mail)
>>Subject: adaptec scsi 2940 on solaris sparc
>>
>>
>>hello,
>>i have a scsi card from adaptec:AHA-2940U2W.
>>I installed it on sun blade 100 with a drive.
>>probe-scsi-all say nothing.
>>When i did
>>ok show-devs, i see something about scsi
>>when i boot, there is only this message:
>>May  2 12:15:20 atarnajuat scsi: [ID 193665 kern.info] sd0 at uata0: 
>>target 1 lun 0
>>
>>but i am unable to use this disk.(with format, or iostat -En).
>>
>>Anybody know if this card works on solaris sparc? IIvisit 
>>adaptec web, 
>>but there is nothing about this card and solaris.
>>
>>TIA,,i will resume.
>>-- 
>>Gerard HENRY
>>_______________________________________________
>>sunmanagers mailing list
>>sunmanagers at sunmanagers.org
>>http://www.sunmanagers.org/mailman/listinfo/sunmanagers
>>
> 
> 
> 



-- 
Gerard HENRY
LATP UMR 6632
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From Charles.Homan at GDC4S.Com  Thu May  2 10:09:21 2002
From: Charles.Homan at GDC4S.Com (Homan, Charles (NE))
Date: Thu, 02 May 2002 10:09:21 -0400
Subject: SUMMARY: DCE nameservice error
Message-ID: <2575327B6755D211A0E100805F9FF9540BB29119@ndhmex02.ndhm.gsc.gte.com>

I received no responses, but starting the "cdsclerk" on the client machine
seems to have resolved the issue.

Regards,
Charles

-----Original Message-----
From: Homan, Charles (NE) 
Sent: Monday, April 29, 2002 1:57 PM
To: 'sunmanagers at sunmanagers.org'
Subject: DCE nameservice error


We are having a problem with DCE.  Unfortunately, the product was purchased
at another site, those people are all gone now, and we have no maintainence.
(I'm working to correct that last problem. :)  I'm hoping one of you kind
souls has some familiarity with DCE and can point me in the right direction.

Our problem is this: one piece of an application under development is
supposed to register with DCE so that other parts can query it.  However,
when we run the server piece we get the error:

  Warning: sec_login_certify_identity returned 387064051
     host configuration info not available (dce / sec)
  Warning: sec_login_certify_identity returned 0
  Error: rpc_ns_binding_export failed in file: /src/Server.cc at line 376.
  DCE Error: Name service unavailable (dce / rpc) 

All I know how to check is that I can do a dce_login.  Does anyone have any
idea what I should be looking at here?

This is a new install (based on sketchy notes from the other site) under
Solaris 7, using NIS & DNS for name services.

It just occurred to me: I believe that the person who installed DCE said
that he had to turn off name services in /etc/nsswitch.conf (that is "hosts:
files" only.)  Could that be a related issue?

Thanks for your help, and I will summarize.

Charles
---
Charles Homan
Systems Administrator
General Dynamics C4 Systems
77 "A" Street
Needham, MA 02494
781-455-2742
charles.homan at gdc4s.com 

The opinions expressed in this e-mail are not necessarily those of GDC4S.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From art at arzoon.com  Thu May  2 12:39:23 2002
From: art at arzoon.com (Art Hebert)
Date: Thu, 2 May 2002 09:39:23 -0700
Subject: Summary: - 16gb of ram in a sun 280r
Message-ID: <83CDD0A27C9AD3118AA1009027AA84FC041485D4@clsrv01.arzoon.com>

Original:

>    I have been reading about 280rs and sun says the max memory is 8gb.
>    Looking at third party memory companies they say it can go to 16gb of
> memory.


Responses:

	They ranged from:

	- No  Problems
	- It will work but sun may not support it
	- Sun will have this in August
	- No way it will work

Here are the responses - I am going to remove the names:

#1
I am a reseller of SUN hardware &  third party memory.  It is my experience
that most servers can handle more
memory than the manufacturers allow.   Additional memory should be a
problem.  However there
many be problems with maintenance.  SUN may not support the server.  Or if
there is a hardware failure of some sort, they may not support it due to the
additional memory.   If its a production box, it may not be a good idea.
If you are using it for test & development,  I'd probably go ahead.

#2
It works from what I understand.  Just keep in mind that Sun service can
refuse service if you're using non Sun memory....

#3
Dataram does offer an 8GB option for the Sun Fire 280R which 
will take your system to a maximum capacity of 16GB instead of 8GB. 
I can give you references if need be. We offer a lifetime warranty on 
our memory. 
#4
The 280R/Sun Blade 1000 has 8 DIMM slots, two banks of four.

I'm not aware of anybody in production with a 2GB DIMM for this
box (not a standard PC133 DIMM).  In fact, I'm not aware of anybody
anywhere with any kind of 2GB DIMM for any box, period.  Not Crucial,
not Kingston, not PC guys like Samsung/Corsair.  I suspect you need
to find smarter third party memory companies, the guys you're talking
to now aren't very good.  :-)


#5
The Sun rep that was here said that currently the Sparc III technology
handles 4Gb per processor (The280R is capable of handling dual processors).
Also, due out in around August the capabilty was for each processor to
handle 8Gb of memory. I would guess the third party memory companies are
looking to the near future or Sun is ahead of schedule. 
2 processors * 4 Gb memory= 8Gb in the Sun literature 
2 processors * 8Gb memory = 16 Gb the third party memory companies are
saying 


Art Hebert		Email: art at arzoon.com <mailto:art at arzoon.com>
Arzoon IT Dept.	Phone: 650 522-6061			Mobile: 650 444-8667
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From walsec at hotmail.com  Thu May  2 14:02:33 2002
From: walsec at hotmail.com (Walse Chen)
Date: Thu, 02 May 2002 14:02:33 -0400
Subject: Summary:where to put SDS state database 
Message-ID: <F171ixjnzi95HfqR77b000000b1@hotmail.com>

Thank you for lot of quick answers.
Thanks to
Donovan Reetz, Karyn Williams, Adam and Christine L, Andrew Caines, Tim 
Chipman, Debbie Tropiano, Larye Parkins, Mike Salehi, Serge Borkov, Darren 
Dunham and Thomas Anders.

My original question was:
Hi admins,

Forgot make a small partiation on root and system disk. The data disks were
mirrored a quit while. I am going to make root&system disk mirror, any goog
idea where to put SDS state databse into? What is the issue if I do not
metadb -a on this two disks.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Solution1:
Almost everyone recommend to steal little bit from swap space. I'll going to 
do this. Delete swap: swap -d /dev/rdsk/c0t0d0s1, creat new small 
partiation, /dev/rdsk/c0t0d0s3, create databas, metadb -a -c 2 c0t0d0s3 
c1t0d0s3. Then swap -a /dev/rdsk/c0t0d0s1.

Solution2:
Since there are several state db on other disks, just go ahead do mirrors 
for root and sytem disk.

Solution 3:
It came from Andrew Caines, he said: If you're feeling brave, you can steal 
the last cylinders on a new and
partly full UFS, then fsck it.*

Thank you guys again

Walse





_________________________________________________________________
Join the worlds largest e-mail service with MSN Hotmail. 
http://www.hotmail.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From MGurugunti at us.nomura.com  Thu May  2 14:31:20 2002
From: MGurugunti at us.nomura.com (Gurugunti, Mahesh)
Date: Thu, 2 May 2002 14:31:20 -0400
Subject: SUMMARY:Weird  find & rm behaviour
Message-ID: <331B688D81962F4E82EC4BA592D4467D1F656A@njex102.us.nomura.com>

Thanks to all who gave inputs. Although most of the folks suggested using
-exec and find command combination, it did not work.

Since the NFS mounted filesystem was from IBM mainframe, I believe it has a
different file structure implementation(MVS file structure implementation).
Hence some incompatibilities.

By trial and error, I was able to find a way to overcome the problem. If the
NFS filesystem from IBM mainframe is mounted with "-noac" (no attribute
caching) option, the same old command worked.(# find . -type f -mtime +14
-print | xargs rm -f) .    -exec and find combination works too.

Here is a note from IBM's support knowledge base which MORE or LESS gives
some hint of the NFS problem.


  8) Symptom: Recursive Remove ( rm -rf ) does not always clean
     up the directory. Users receive "Directory not empty"
     messages and may need to issue the command again.
     Due to the uniqeness of MVS file architecture, it is not
     possible to make rm -rf work the same as Sun. Development
     tested and confirmed that this command does work with AIX
     using V2 NFS protocol.


Mahesh




>  -----Original Message-----
> From: 	Gurugunti, Mahesh  
> Sent:	Monday, April 29, 2002 12:05 PM
> To:	'sunmanagers at sunmanagers.org'
> Subject:	Weird  find & rm behaviour
> 
> I am trying to remove the files which are 14 days old as below, when I
> used the combination of find and rm looks like it is removing only one
> file every time I run and not all the files.
> 
> The /usr/App/Ntaps/Trans_data is NFS mounted from a IBM mainframe. 
> 
> Any, clues as to why this is happening?? 
> 
> TIA
> Mahesh
> 
> 
> # pwd
> /usr/App/Ntaps/Trans_data
> # find . -type f -mtime +14 -ls
> 18171  148 -rw-rw-rw-  1 lb       ingadm     150300 Apr 12 21:45
> ./ntaps_daily_cust_acct.0412
> 18143  180 -rw-rw-rw-  1 schepis  users      183366 Apr 11 22:08
> ./ntaps_daily_cust_acct.04120730
> 18152  148 -rw-rw-rw-  1 schepis  users      150300 Apr 12 21:49
> ./ntaps_daily_cust_acct.04150730
> 18123    8 -rw-rw-rw-  1 schepis  users        4509 Apr 11 14:30
> ./ntaps_inact_cust_acct.04111436
> 18126    8 -rw-rw-rw-  1 schepis  users        4509 Apr 11 15:30
> ./ntaps_inact_cust_acct.04111536
> 18132    8 -rw-rw-rw-  1 schepis  users        4509 Apr 11 16:31
> ./ntaps_inact_cust_acct.04111637
> 18136    8 -rw-rw-rw-  1 schepis  users        4509 Apr 11 17:00
> ./ntaps_inact_cust_acct.04111703
> 18153    4 -rw-rw-rw-  1 schepis  users        1002 Apr 12 17:00
> ./ntaps_inact_cust_acct.0412
> 18144    8 -rw-rw-rw-  1 schepis  users        4509 Apr 11 22:08
> ./ntaps_inact_cust_acct.04120730
> 18172    4 -rw-rw-rw-  1 schepis  users        1002 Apr 12 21:49
> ./ntaps_inact_cust_acct.04150730
> 18159    8 -rwxrwxrwx  1 lb       ingadm       6567 Apr 12 17:02
> ./tsy_mark.dat.14998
> 18160    4 -rwxrwxrwx  1 lb       ingadm       1142 Apr 12 17:04
> ./tsy_mark.dat.15426
> 18137    8 -rwxrwxrwx  1 lb       ingadm       7652 Apr 11 17:02
> ./tsy_mark.dat.8637
> 18148  676 -rw-rw-rw-  1 lb       ingadm     691405 Apr 12 03:59
> ./CLEAR/ntaps_clear.flat.used.020412
> 
> 
> # find . -type f -mtime +14 -print | xargs rm -f
> rm: cannot change to /mbs/Trans_data directory: No such file or directory
> 
> 
> # find . -type f -mtime +14 -ls
> 18143  180 -rw-rw-rw-  1 schepis  users      183366 Apr 11 22:08
> ./ntaps_daily_cust_acct.04120730
> 18152  148 -rw-rw-rw-  1 schepis  users      150300 Apr 12 21:49
> ./ntaps_daily_cust_acct.04150730
> 18123    8 -rw-rw-rw-  1 schepis  users        4509 Apr 11 14:30
> ./ntaps_inact_cust_acct.04111436
> 18126    8 -rw-rw-rw-  1 schepis  users        4509 Apr 11 15:30
> ./ntaps_inact_cust_acct.04111536
> 18132    8 -rw-rw-rw-  1 schepis  users        4509 Apr 11 16:31
> ./ntaps_inact_cust_acct.04111637
> 18136    8 -rw-rw-rw-  1 schepis  users        4509 Apr 11 17:00
> ./ntaps_inact_cust_acct.04111703
> 18153    4 -rw-rw-rw-  1 schepis  users        1002 Apr 12 17:00
> ./ntaps_inact_cust_acct.0412
> 18144    8 -rw-rw-rw-  1 schepis  users        4509 Apr 11 22:08
> ./ntaps_inact_cust_acct.04120730
> 18172    4 -rw-rw-rw-  1 schepis  users        1002 Apr 12 21:49
> ./ntaps_inact_cust_acct.04150730
> 18159    8 -rwxrwxrwx  1 lb       ingadm       6567 Apr 12 17:02
> ./tsy_mark.dat.14998
> 18160    4 -rwxrwxrwx  1 lb       ingadm       1142 Apr 12 17:04
> ./tsy_mark.dat.15426
> 18137    8 -rwxrwxrwx  1 lb       ingadm       7652 Apr 11 17:02
> ./tsy_mark.dat.8637
> 18148  676 -rw-rw-rw-  1 lb       ingadm     691405 Apr 12 03:59
> ./CLEAR/ntaps_clear.flat.used.020412
> # 
> 
> 
> At the end of second operation, it removed only the file with the
> timestamp Apr 12 21:45.
> 
> 
> Mahesh
> Nomura Securities International
> WFC 2 , Bldg B 18th Floor
> NYC, New York 10281
> 
> 
-----  This message is for the named person's use only. It may contain
confidential, proprietary or legally privileged information. No
confidentiality or privilege is waived or lost by any mistransmission. If
you receive this message in error, please delete it and all copies from your
system, destroy any hard copies and notify the sender. You must not,
directly or indirectly, use, disclose, distribute, print, or copy any part
of this message if you are not the intended recipient. Nomura Holding
America Inc., Nomura Securities International, Inc, and their respective
subsidiaries each reserve the right to monitor all e-mail communications
through its networks. Any views expressed in this message are those of the
individual sender, except where the message states otherwise and the sender
is authorized to state the views of such entity. Unless otherwise stated,
any pricing information in this message is indicative only, is subject to
change and does not constitute an offer to deal at any price quoted. Any
reference to the terms of executed transactions should be treated as
preliminary only and subject to our formal written confirmation.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From ltiu at alumni.sfu.ca  Thu May  2 22:24:13 2002
From: ltiu at alumni.sfu.ca (ltiu)
Date: Thu, 2 May 2002 19:24:13 -0700
Subject: Summary: SS5 uses only 32MB out of 128MB Memory
Message-ID: <20020503022236.CKRO23644.priv-edtnes16-hme0.telusplanet.net@there>

Hello,

For some reason that I think is normal with Solaris2.6/SunOS5.6:

It tries it's best to use less than 32MB RAM at all time and will only use 
the rest of the RAM when you are really stressing it - like gunzipping a 
150MB file or running a Java program that uses up 128MB RAM. Weird. I do not 
see this behavior in Solaris8.

Thanks for all you guys who suggested solutions.

ltiu



Subject: SS5 uses only 32MB out of 128MB Memory
Date: Wed, 1 May 2002 20:48:59 -0700
From: ltiu <ltiu at alumni.sfu.ca>
To: sunmanagers at sunmanagers.org

Hello guys,

Found an interesting problem that I did not notice for over 3 months.

The SS5 that I am using uses only 32MB of RAM, regardless of whether there is
64, 96 or 128MB of total RAM installed. My system uses swap all the time
since 32MB that it use is not enough. The funny thing is that top reports
that a lot of free RAM is available but it never gets used even during heavy
memory usage like starting up iPlanet.

I've tried switching the RAM around between memory slots but this did not
solve the problem.

What could be the problem here?

ltiu
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers

-------------------------------------------------------
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From erwin.mendoza at cronusgroupinc.com  Fri May  3 06:43:48 2002
From: erwin.mendoza at cronusgroupinc.com (Erwin B. Mendoza)
Date: Fri, 3 May 2002 18:43:48 +0800
Subject: SUMMARY: PATH for MAN Page
Message-ID: <!~!UENERkVCMDkAAQACAAAAAAAAAAAAAAAAABgAAAAAAAAAjPjS2Ckl1hGgawBAlTRo/MKAAAAQAAAAGVbBLLRe1hGgawBAlTRo/AEAAAAA@cronusgroupinc.com>

Hello everyone,

Most of the replies are advised me to set it in my .profile or
/etc/profile...... Actually, from my /etc/profile, it doesn't have the
defined MANPATH, so I've viewed it from my environment variables the
original path for the MANPATH and then added it to my /etc/profile
because it is our global profile........ Someone also suggested that try
to run the command "find / -name man -type d -print 2>/dev/null", so
what I did is defined all the directories that had manual pages........

Anyway, many thanks to all who replied to my queries especially to:
Lennon, Padraig 
Pierre Zimmermann 
Glass, David 
zoo11 at mail.hst.co.kr
Tan Tshun Kiat 
Toby Gollop 
Anthony Worrall 
Spencer Hoffman 
robert.forsey at axa-sharedservices.co.uk
SteinAxt 
Martynas Buozis 
ed.rolison at itc.alstom.com
Martin Carpenter 
Schmitt, Martin
Patrick Li
Geoff Reed

and also to moderator of this group......

Thanks and more Power.....

Regards,
ERWIN


BELOW IS MY QUESTION.......

I have installed some packages in my Solaris Box. I'm trying to use the
MAN page for an application but I've received "No manual entry for
....".

How will I put it in my PATH? And also, could you give some directory
list where I can see some MAN pages that are not included in the
PATH..... I've seen that there's some directory located in my
/usr/local/man, is this the only one who are not in my automatic MAN
page in PATH?......

Thanks and more power.......

 Best Regards,

 Erwin B. Mendoza _______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From min at psde.mec.mei.co.jp  Fri May  3 09:54:17 2002
From: min at psde.mec.mei.co.jp (Min Oo Tint)
Date: Fri, 3 May 2002 14:54:17 +0100
Subject: SUMMARY: perfmeter and what are the measuring units and more
  and  LSF
Message-ID: <3C667B62E467F44DB51E01DC328A3431136284@psde-server>

Thanks for all who replied:

Paul Greidanus [paul.greidanus at ualberta.ca]
Russ Poffenberger [poffen at san-jose.tt.slb.com]
Kevin Buterbaugh [Kevin.Buterbaugh at lifeway.com]
Jay Lessert [jayl at accelerant.net]
Michael Knox [michaelk at icpdd.neca.nec.com.au]


Yes, only one CPU is being used.

The application would have to be written with threading in mind to be
able to use more CPU's.

You can run other processes/applications at the same time and it will spread
them among the CPU's.


If your application is single-threaded, then there's no way to make it use
both CPU simultaneously.  Of course, as it is interrupted it may be switched
from CPU to CPU, but it'll never use more than one of them. Having 2 CPUs is
still good; it allows "system" processes like sched, fsflush, etc. to run on
one CPU without having to preempt your process running on the other CPU.  It
does sound like the 2nd CPU is going mostly unused, however.

Yes, AFAIK you can use LSF to schedule batch jobs on the other processor.
Good luck...


Ah yes.  You mean the VCS Verilog simulator from Synopsys, correct?

As far as I know, there is no such thing as a multithreaded Verilog
simulator *anywhere*.  It is a very difficult problem.

> How can I find out what the second processor is doing?

You can run prstat.  What you'll find is that your VCS process is actually
moving between the two processors (doesn't matter, since it's constantly
refilling the L2 cache anyway).  The "other processor" (whichever one is not
running VCS at the time) is handling I/O, interrupts, running cron and nscd,
etc., etc.  Basically idle.

> and 
> can I used LSF to direct batch job(s) to say 2nd processors?

Of course.  If you've got enough RAM.

That is the only practical way to speed up Verilog with multi-processors, by
the way.  You break up a single long simulation into multiple shorter
simulations and run them in parallel.


Most Synopsys tools are single threaded, hence you will only use 1 CPU  for
each job. The other is basically handling the other processes on the box. If
you start another Synopsys job on this box it will use the 2nd CPU.

LSF is only useful to you if you want to queue jobs based on various
resource parameters including license availablity. and recommended that I
used



mpstat 	- 	is a good tool... it gives per CPU information..

sar, vmstat, mpstat, iostat, and netstat 	-	to monitor your
system.  


My question was:

Hi,

I am running VCS from Synopsys on Sun Blade 1000, 2 CPU, using perfmeter to
Measure the performance of the system, from the log I see

29/04/2002 18:38:41 einstein    cpu = 0.00      load = 0.00     disk = 0.00
intr = 215.00   pkts = 4.00

Can someone tell me what they are measuring and what are the units?

I plotted these on the graph and found that 50% of cpu is fully used Through
out the simulation period. Am I to assume that only one of the processor is
being utilized and if so, any way I can spread this load evenly on both
processors?

Also I have been told that VCS is single threaded hence could this be the
reason?

How can I find out what the second processor is doing? and can I used LSF to
direct batch job(s) to say 2nd processors?


Regards,


Min Oo Tint

Matsushita Electric Europe
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From mikelist at sky.net  Fri May  3 11:04:49 2002
From: mikelist at sky.net (Mike's List)
Date: Fri, 3 May 2002 10:04:49 -0500 (CDT)
Subject: SUMMARY: SUNWadmap
In-Reply-To: <Pine.GSO.3.96.1020502173955.1199B-100000@pulsar.sky.net>
Message-ID: <Pine.GSO.3.96.1020503095527.9148B-100000@pulsar.sky.net>

Thanks goes to Dan Astoorian, I needed to add SUNWadmc, no more error
message and the sys-unconfig/sysidconfig actually works :)


- Mike


bash-2.03# pkginfo -l SUNWadmc
   PKGINST:  SUNWadmc
      NAME:  System administration core libraries
  CATEGORY:  system
      ARCH:  sparc
   VERSION:  11.8,REV=2000.01.21.04.37
   BASEDIR:  /
    VENDOR:  Sun Microsystems, Inc.
      DESC:  Core software libraries used for system administration
    PSTAMP:  on28-patch20011128153101
  INSTDATE:  Apr 14 2002 13:29
   HOTLINE:  Please contact your local service provider
    STATUS:  completely installed
     FILES:     29 installed pathnames
                 4 shared pathnames
                 4 directories
                 1 executables
              3557 blocks used (approx)



On Fri, 3 May 2002, Mike's List wrote:

> Solaris 8, core installation, I added SUNWadmap to config/unconfig quicker
> but now everytime the system booted I received the below...sysidnet and
> sysidconfig is there, am I missing some other package for sysidnet and
> sysidconfig?
> 
> ld.so.1: /usr/sbin/sysidnet: fatal: libadmutil.so.2: open failed: No such
> file or directory
> Killed
> 
> ld.so.1: /usr/sbin/sysidconfig: fatal: libadmutil.so.2: open failed: No
> such file or directory
> Killed
> 
> When adding SUNWadmap, the system indicates that SUNWmfrun and SUNWctpls is
> required and I installed those two packages (I think it's mainly for GUI).
> 
> Thanks.
> 
> 
> - Mike
> _______________________________________________
> sunmanagers mailing list
> sunmanagers at sunmanagers.org
> http://www.sunmanagers.org/mailman/listinfo/sunmanagers
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From sunmanagers at cs.du.edu  Fri May  3 17:55:52 2002
From: sunmanagers at cs.du.edu (Ivan Fetch)
Date: Fri, 3 May 2002 15:55:52 -0600 (MDT)
Subject: Summary: Documenting infrastructure
In-Reply-To: <20020502052133.12948.qmail@kev.wpafb.af.mil>
Message-ID: <Pine.GSO.4.44.0205031506030.21527-100000@everest.cs.du.edu>

Hello,
   I'd like to thank the following for replying to my inquiry about
documenting one's Unix infrastructure - I'm sorry if somehow I've missed
anyone:
 Jason Husk <Jason.Husk at corp.infi.net>,
Richard Grace <rgrace at aapt.com.au>,
"Homan, Charles (NE)" <Charles.Homan at GDC4S.Com>,
Greg Gallagher <ggallag at foc.com>,
gene.g.beaird at mail.sprint.com,
Harvey Wamboldt <harvey at iotek.ns.ca>,
"Thomas M. Payerle" <payerle at physics.umd.edu>,
Jay Lessert <jayl at accelerant.net>,
Jim Southerd <jsoutherd at bakersfield.com>,
Bertrand_Hutin at notes.amdahl.com,
Andrew_Rotramel/CCHLIS at cch-lis.com,
Karl Vogel <vogelke at dnaco.net>

   I have included all of the replies below because I feel this is a bit
more dynamic of a discussion than "how do I get my ABC to XYZ?" - I hope
the folks who replied don't mind their messages getting re-posted to the
list.

The gist of the methods of documentation and things being documented
include:
* a directory structure of text files (this is currently what we have)
accessable from local filesystems and/or the web, consisting of various how-to notes
   * homemade man pages outlining common sys admin tasks and user
frequently asked questions
   * scripts which compile server info (i.e. disk partition layout) into a web tree along with
typescripts of installed software and other How-Tos
   * using xml+DocBook to publish printed and online sys admin docs
   * a combination of docs in databases, MS Excel, and JPG images dealing
with inventory, machine room layout, Etc
   * various html pages documenting "internal" procedures and How-Tos, as
well as instructions for end-user consumption

   Other good points worth summarizing here are: The more trouble it
is to update docs (i.e. work must be done on a specific machine or
specialized software is required) the less likely it is that they will be
updated as frequently as you'd like.  Successor sys admins will most
likely only want to read docs up to a certain point, and will fall back on
them if/when they're at a loss for how something works or why it is
configured in such a way down the road.  The more docs you have, the more
there is to maintain (we may end up writing more docs than
fixing/improving things).

   My game plan at this point is to continue playing with xml+docbook (my
most favorite suggestion), latex, and texinfo to see which yields the best
output in the desired formats (something like postscript for printing and
text and html for online versions).  I think I can modularize the docs
enough (each sub-section living in a separate file) that maintinance is
minimally daunting.  I'll either use Excel spreadsheets or some other
database driven solution to store inventory (model, serial number, RAM,
disk, repair history) which can be imported into the xml/latex document
easily enough (i.e. CSV export).  I like the idea of pictures of the
server room to give people something to look at and associate with while
they're reading some
particular doc.

   I'll post another message as things are progressing into something
meaningful -- Thanks very much for all the feedback!


Ivan Fetch.


 Here is my original post:
> Hello all,
>
>    I'd like to start maintaining more documentation designed for
> myself (reminders), folks who are replacing me (when that day comes), and
> anyone needing to respond to an emergency if I'm unavailable.  I invition
> including such things as where non-packaged software is installed on our
> servers, brief How-Tos on typical tasks (i.e. restoring backups, user
> maintinance, jumpstarting new workstations), caveats on upgrading
> frequently used software which gets patches often (including Solaris), the
> organization of our IP address space, the configuration of our printers
> (jet-direct -> Solaris -> Samba -> Windows clients), et cetera.
>
>    I'm writing to solicit other people's opinions on the creation of
> something like this - if you were starting a position, what kinds of
> things would be helpful to have some background on vs. figuring things out
> on one's own?  How are you folks out there documenting your infrastructure
> (as much for yourselves as for future employment)?  Do most of you use
> something like rcs or cvs to manage your configuration files (/etc,
> /usr/local/etc)?  What do folks use to type set electronic and printed
> documentation -- I've been thinking latex or sgml -- I would like
> automatic referencing (being able to refer to sections by number, title, and/or page number) as well as an automatically
> generated table of contents.
>
>    I appreciate any input you all are willing to give  - I'll definitely
> summarize.  I'd like to start converting the various notes I've made while
> reconfiguring things into a more organized and useful structure.
>
> Thanks,
> Ivan Fetch.

----- responces -----


On Wed, 1 May 2002, Jason Husk wrote:

>
> 	Something that I used to use when I helped maintain such an
> archive is a simple man page.  What we did was we had a rather simple
> utility that would take a plain ascii text file and convert it to a man
> page and then place it into a common directory that we used to store the
> docs in.  This directory had only group permissions on it so we had to be
> in a member of the admin group to be able to read this directory and the
> info docs contained in it.  Then if we needed to look something up all we
> had to do is make sure that directory was in our MANPATH environment
> variable and then we could access the pages via the built in man
> command... ie man boot-net .... or man NT_printers ... you get the idea I
> hope.  It was very helpful for when I was just starting out.
>
> 	We had items such as how we setup user accounts, changed thier
> passwords, located source files that we compiled for public use, how we
> built various types of workstations.  Items such as common user questions
> were stored too like compiler and library locations.  Other items like
> rules pertaining to how we allowed our users to use the computers and
> access the internet.  (It was a university setting).  Also common everyday
> tasks were listed to things that an end user might ask that the sysadmin
> may not necessarily need to know for their every day work.  How to setup a
> public_html directory for a  web page.  How to setup a .htaccess file.
> How to submit grades etc.. These may be some helpful docs to have lying
> around if you have spare time to create them.
>
>
> Anyway I wont ramble on too much but I hope it helps


On Wed, 1 May 2002, Richard Grace wrote:

> Hi there.  What a task, eh?
>
> I use scripts to dump disk, network, memory, and other information
> into a temporary directory tree on a nightly basis.  This is triggered
> by a script which sshs to each server, executes the script, tars up
> the directory, copies and untars into our web tree.  The files are
> linked by indexes in the web tree.  This is the easiest way I've found
> to keep online information on running servers.
>
> We have a similar web tree structure which links to the jumpstart
> config and packages area.
>
> Most of the software we use is packaged, and there are READMEs
> in each package which (should) detail the last configuration options
> used for the installation, and the output from script, as well as the
> source tarball.  I know this wastes a lot of disk space, but it is more
> thorough that way.
>
> We also have a "links" page, which links to as many howtos as the
> operators have found useful in the past.
>
> Richard.


On Wed, 1 May 2002, Homan, Charles (NE) wrote:

> At my current employer, they had a simple text file setup called
> "miscnotes".  Things in it are arranged in sections (by application, mainly)
> and include notes on how things were installed, how to add users, notes on
> special hardware, standard disk partitioning, bugs encountered and their
> workarounds, license info, rarely-used but necessary administrative command,
> etc.  Longer notes are in seperate files in the same directory, with a
> reference in miscnotes.
>
> This has had numerous advantages over "fancier" (i.e.: laid out and/or
> printed) documentation:
>
> o completely searchable (via "grep -i") and human-readable
> o always where I need it (i.e.: on the server(s))
> o can cut and paste lines to command line for seldom-used commands with long
> syntax (read: NIS+)
> o simple to keep updated
>
> This last point is key to me.  I would much rather have an accurate and
> up-to-date text file than a well-laid out, nice looking piece of
> documentation that someone created two years ago and eventually stopped
> maintaining 18 months ago.  KISS. ;-) That said, some things (I'm thinking
> of network layout here) can be much more easily represented (and understood)
> graphically, and others (like inventories) are more useful in spreadsheet
> form.  Use the right tool for the job, but make sure everything is in a
> "findable" place.
>
> Hope this helps!
> Charles


On Wed, 1 May 2002, Greg Gallagher wrote:

> Quite nice of you to take this initiative -- preparing your successor
> and other junior admins, etc.  Not many SA's do!  :-)
>
> It's a little more troublesome, but I'd suggest looking into XML +
> Docbook and organize a CVS repository where you can build an "Admin
> Guide".  I did this at a previous job, and it turned out really cool.
> By the time I left, I had written a little chapter on all my major
> duties such as backups and restores, installing Oracle, etc.  It was a
> perfect handbook to give the poor bloke who took my job :-)
>
> The thing about using XML (or SGML if you're brave enough) is that
> you'll be able to publish it to the web easily, as well as making a
> printable manual.  It's worth the time, and once you have a basic
> structured document setup it's quite easy to keep it up to date.
>
> cheers,
> --
> Greg Gallagher
<followup>

> On Wed, May 01, 2002 at 07:16:56AM -0600, Ivan Fetch wrote:
> > Hi,
> >    Thanks very much for your reply - Might you have any pointers RE:
> > xml+docbook (did this take place under Solaris when you worked on it)?
.....
> Hrm, well.  At the time I was using Linux and actually it was SGML +
> Docbook using openjade.
>
> Now-a-days, you could look at http://www.xmlsoftware.com for some
> software.  Anything java should work on Sun.  Docbook you can get from
> http://www.docbook.org, or a pointer thereoff.  I'd *highly* suggest
> using Emacs + PSGML mode for doing the actual editing and
> composition.  That works good for me.
>
> Also, you could use XSLT to do the transforming.  I think that works
> the best.  Then use SAXON to do the transformation of the XML +
> Docbook XSL to HTML or whatnot.
>
> Take a look at this page, it looked really useful to me:
>
> http://www.dpawson.co.uk/docbook/
>
> Good luck!
>
> Oh, another idea, I just thought of was to use something like Apache
> XML to do transformation on the fly.  There are a few free options to
> do that type of thing, and a few really expensive Commercial products
> as well (such as Documentum and Arbortext).
>
> cheers,
> --
> Greg Gallagher


On Wed, 1 May 2002 gene.g.beaird at mail.sprint.com wrote:

> Ivan,
>
> Please post your results, I am VERY interested.  That said, I have been
> using various things to help with a 'Greenbook' for some time.  I have
> used things as simple as text files and Word documents to databases.
> It seems that for me, at least, no one tool does it all (unless you get
> one of those multi-thousand-dollar monstrosities that are commercially
> available that try to be everything for everyone all the time, but
> really don't do anything well).
>
> I have always tried to keep it as low-tech and generic as possible.
> When I was maintaining a Macintosh network at an oil company, I kept
> everything on a Filemaker Pro database. That way, I could slice and
> dice the information any way I wanted to keep whoever wanted to see it
> happy (or mostly happy).  If I had walked into this present job knowing
> then what I do now, I would have converted their documentation to some
> sort of database, or another.  That way, you can be sure you are
> maintaining the same data on all the systems you have to maintain.  It
> makes things like machine room diagrams, network diagrams and general
> comments difficult, but dome database applications can even handle
> this.
>
> We use a series of Word documents, Excel spreadsheets, jpeg images
> (photos of the server room equipment) and Visio documents.  That way,
> almost anyone in my group can look at the documentation and use it.
> You still, IMHO, need to keep it simple.  If it runs under some
> behemoth app that is difficult to install and use, you will have
> difficulty getting others to document their work.  It will be equally
> difficult to use as you can bet a machine that has the proprietary
> software installed will not be available when you need it most.
>
> I have assembled a 'Greenbook' or 'IT Operations Manual' using a set of
> loose-leaf notebooks.  That way printed material is available for quick
> reference and it is easy to update.  Keep NO passwords in the
> Greenbook, but get pretty specific.  We also control who gets our
> Greenbook, and limit it to the IT staff and their immediate managers.
>
> I have a standard 'Information Sheet' in M$ Word format, if you would
> like to see it.
>
> Regards,
>
> Gene Beaird
> Systems Integrator V


On Wed, 1 May 2002, Harvey Wamboldt wrote:

> Interesting idea.  I'd be very interested in reading your summary.
>
> We have a very small shop, and often I'm called on to "consult" on
> problems when I'm off site.  I have a simple directory full of ASCII
> text files that contain my notes.  If I can get on the Internet I can
> access these whether I'm on the east coast or the west coast.  Our
> other SysAdmin (who's primarily a PC person) knows where to find these
> notes.  I use RCS to maintain versions, but that's more for security
> against accidentally deleting a file (so I don't have to retrieve it
> from the backup).  I use rsync to maintain copies on my home PC.  The
> biggest problem is outdating obsolete information.  Mostly that
> doesn't happen since I don't take the time.  I also have a public_html
> directory where I keep some information, primarily for users.  I write
> the HTML by hand, and try to keep it brief.  Oh yes, I find every
> directory needs an INDEX file and a README file.  It's amazing what a
> difference that makes.
>
> Best Rgds,
>
> -H-
>


On Wed, 1 May 2002, Thomas M. Payerle wrote:

> On Tue, 30 Apr 2002, Ivan Fetch wrote:
>
> > Hello all,
> >
> >    I'd like to start maintaining more documentation designed for
> > myself (reminders), folks who are replacing me (when that day comes), and
> The dark area of sys admin.
>
> As for how much to document, I say can never have too much documentation
> (albeit, you can waste too much of your time documenting).  Basically, the
> new person will decide for himself what he wants to learn from documentation
> and what to discover on his own, and may decide to refer back to documentation
> previously ignored if he discovered something nasty.
>
> The only down side of a lot of documentation is that there is more stuff to
> keep up to date, or as usually is the case, there is more out of date
> documentation.  Not that out-of-date documentation is truly evil (it usually
> has merit for at least understanding historical significance of decisions,
> and usually is applicable with minor changes.  Reader just needs to know that
> cannot blindly follow documentation).
>
> I try to use RCS to track changes to assorted config files, not always very
> successful (not infrequent to need to change something only to find already
> checked out and don't know what last change was, despite being me who checked
> it out).  But at least are able to quickly revert to working versions, and
> beats 500 files of /etc/inetd.conf.sav.20020501 and ilk.
>
> As for format of documents, we generally do html.  Post on a password protected
> page (nothing is really confidential, but don't want users stumbling upon it
> and getting enough rope to hang themselves), and duplicate on a second web
> server, as well as printed copies.  This is mainly because html is most widely
> readable format after ascii, and being on web stresses the ephemeral nature of
> the contents.
>
> I personally like latex, but fewer and fewer people who know how to use, and
> most of the documentation (here at least), is out of date by the time it hits
> the printer, so we prefer on-line (also out of date, but who is stupid enough
> to take as gospel something got off the web:)
>


On Wed, 1 May 2002, Jay Lessert wrote:

> On Tue, Apr 30, 2002 at 11:10:49PM -0600, Ivan Fetch wrote:
> >    I'm writing to solicit other people's opinions on the creation of
> > something like this - if you were starting a position, what kinds of
> > things would be helpful to have some background on vs. figuring things out
> > on one's own?
>
> I've been through this recently/currently; walked into somebody else's
> shop about a year ago, some things were documented, some weren't, I'm
> writing documentation as i go.
>
> 1) How to create a new account.  For us, NIS/NTDC/qmail aliases/phone
>    setup.
> 2) How to add a new host.
>
> Those two are *really* nice to have.
>
> > How are you folks out there documenting your infrastructure
> > (as much for yourselves as for future employment)?
>
> Mostly in HTML.  We used to use FrameMaker (which we already use here
> for product datasheets and app notes), but it's not worth it for
> short simple documents like these.
>
> > Do most of you use
> > something like rcs or cvs to manage your configuration files (/etc,
> > /usr/local/etc)?
>
> RCS (again, keep it simple).  We use RCS for the HTML systems docs as
> well.
>
> --
> Jay Lessert                               jay_lessert at accelerant.net


On Wed, 1 May 2002, Jim Southerd wrote:

> I'm in the same process.
>
> The first thing I have documented is my disk structure on the array for both of my servers.
> I done this by printing the structure right from Veritas from each system, and superimposing some relevant info for our application, along with some pertinent info of each server.
>
> Now I'm 'better' documenting all my scripts and printing them out, along with the printout of the crotab.
>
> Next will be the file structure from the application perspective.
>
> Then we'll see.
>
> Our Director used Visio to document the entire network from information given to him from each sysadmin.
>
> Now he and several of us, are working on documenting the various automated (and manual) ftp's that are going on between servers.
>
> Hope this helps!
>


On Thu, 2 May 2002 Bertrand_Hutin at notes.amdahl.com wrote:

>
> In the days were I was a sys admin, I gathered all the workstation
> description as a set of html pages.
> I was using rcs for some config files and also had copies of some in an
> specific (non-root) account home.
> I had also written a manual (LaTex) to be used by people when i was on
> vacation.


On Thu, 2 May 2002 Andrew_Rotramel/CCHLIS at cch-lis.com wrote:

>
> I am doing this now, but because of the office culture, have to use Word
> for the book. My goal is to document things such that if I am squashed by a
> meteor, someone else can step right into my shoes. I am not worried about
> job security, and think that if anyone tries to secure their job by doing
> secret configurations, they should be fired.
>
> Andrew
>


On 2 May 2002, Karl Vogel wrote:

> I> Do most of you use something like rcs or cvs to manage your
> I> configuration files (/etc, /usr/local/etc)?
>
>    Yes.  I have it done automatically every night.
>
> I> What do folks use to type set electronic and printed documentation --
> I> I've been thinking latex or sgml -- I would like automatic referencing
> I> (being able to refer to sections by number, title, and/or page number)
> I> as well as an automatically generated table of contents.
>
>    You might like texinfo, it can do all of the above and it's a lot more
>    readable than latex.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From kdc at rochester.rr.com  Sat May  4 00:44:37 2002
From: kdc at rochester.rr.com (Kevin Colagio)
Date: Sat, 4 May 2002 00:44:37 -0400
Subject: Summary: Changing system's identity without a reboot....
Message-ID: <001301c1f326$657137b0$021710ac@damnthing>

Thanks to:
Sean Quaint
Alex Stade
Jed Dobson
Eric Shafto
Thomas Lester
Moti

Here are the files that need to be changed:

/etc/hosts
/etc/nodename
/etc/net/ticots/hosts
/etc/net/ticlts/hosts
/etc/net/ticotsord/hosts
/etc/hostname.* (as needed if multiple interfaces)
/etc/resolv.conf (as needed)
/etc/netmasks (as needed)
/etc/domainname (as needed)

NIS/LDAP/NIS+/etc...maps (as needed)


Run these commands:

uname -S new.host.name
hostname new.host.name
ifconfig {reconfiguration information} (as needed)
mkdir /var/crash/`hostname` (just in case)
dumpadm (to update /etc/dumpadm.conf)
/usr/bin/domainname {newdomain} (as needed)


Restart these services...

/etc/init.d/rpc
/etc/init.d/sendmail
/etc/init.d/dtlogin
/etc/init.d/nfs.server (if applicable)

This should cover it all....there were many duplicates of info, so this
is the consolidated list.

Thanks everyone!

Kevin Colagio
kdc at rochester.rr.com
System Administrator and Perpetual Student.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From san_jais at eudoramail.com  Sat May  4 22:09:33 2002
From: san_jais at eudoramail.com (sanjim jaike)
Date: Sat, 04 May 2002 19:09:33 -0700
Subject: Summary:Deleting files with strange names
Message-ID: <ODCPHCNFKDGEBAAA@shared1-mail.whowhere.com>

Hi ,
  Thanks to all those who replied.Special thanks to Dave Mitchell who provided the solution first.The solution by Jeff Woolsey also works,thanks for reply.

Sanjeev

Original Problem:

> Hi,
>   I am trying to delete files with strange names like quotes ,spaces or backslashes.
> 
> find /dir -name '*.inactive' -print | xargs rm
> 
> This command fails when xargs encounters a control character or a quote.



Solution from Dave Mitchell:

If you have Perl installed, the following should do the trick (as long as
the filename doesnt have a \n it it).

    find . -name '*.inactive' -print | perl -ne 'chomp; unlink'

or if you want error reportting too:

    find . -name '*.inactive' -print | \
	perl -ne 'chomp; unlink or warn "Failed to delete $_: $!\n"'

Solution from Jeff Woolsey:

That's a problem with xargs.  The interface does leave a little to be desired,
though it's better than the shell ( rm *.inactive  with special characters).

Try  find2perl  with the same arguments (not the pipe, though), and edit the
resulting perl code to do what you want.

Or use somethig like this:

#! /usr/local/bin/perl

###     xargs rm without the globbing.
#
#       Hmm.  It's not this simple.

@_ = <>;
chop(@_);
unlink @_;
exit 0;




Join 18 million Eudora users by signing up for a free Eudora Web-Mail account at http://www.eudoramail.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From min at psde.mec.mei.co.jp  Sun May  5 13:58:27 2002
From: min at psde.mec.mei.co.jp (Min Oo Tint)
Date: Sun, 5 May 2002 18:58:27 +0100
Subject: SUMMARY: Should NIS master be also NIS client?
Message-ID: <3C667B62E467F44DB51E01DC328A343113628C@psde-server>

Thanks for all those who replied:

Angel Rivera 
Tony Tran 
Jennifer Stults
Moti
David Foster
Vahid Moghaddasi
Hichael Morton
Dave Anderson
system administration account [sysadmin at astro.su.se]

I could say that I did every thing right in getting NIS up and except to
check the file, /etc/nsswitch.conf was correct. After doing sys-uncofnig the
system renamed the file to /etc/nsswitch.nis. Replacing /etc/nsswitch.conf
/etc/nsswitch.nis resolved the problem I had.

Many thanks to  all who replied.

Regards,


Min Oo

------------------------------------

* NIS master normally is also a client, make sure that you view/modify
/etc/nsswitch.conf to reflect NIS usage and make sure exported FS are
mounted on different location than they are being exported from so you wont
craete a deadlock.

------------------------------------

* But I'm not sure whether the NIS maps are preserved after a sys-unconfig,
look in /var/yp/`domainname` to check. 

* Master and client is just conceptual. master is the one that is providing
the service (NIS service) client is the one that is getting the information
from the master. Now if you are logged into the NIS server , and run the YP
command: "ypmatch passwd min" to find out what your NIS passwd is for
example, you are really requesting a service, therefore this nis master is
now a client

* Your nis server should be a client as well , unless you dont need the nis
info it's distributing

* don't forget to set /etc/defaultdomain

* If you want the machine that is the NIS master server to be able to use
NIS, you will need to make it a client of itself or an NIS server.  Sun says
it is possible to make the machine that is the NIS master server a client of
an NIS slave server and not compromise the NIS master server processes.

* When a system is configured to run NIS services, it does not necessarily
mean it needs to consume the same NIS services it provides.  Thus, the
client and the server are in two pieces.  The "ypinit",
/usr/lib/netsvc/yp/ypstart and ypstop scripts and /etc/rc.* scripts assume
if your running a slave or server you need to start both the server and the
client pieces up.In most cases, yes, the NIS/YP master and slaves run both
the server and the client. The biggest reason not to setup the NIS master as
a client is if you move the NIS files to /var/nis or /var/nisfiles and do
not let the general public login to the server host.  You need to modify a
few files, namely /etc/init.d/nis.client and /var/yp/Makefile, but some
higher security environemnts like this.  The general I/T does not know how
to deal with it, so it is generally not done.

* Yes, you need to run ypbind on any machine that needs to query the NIS
maps, even if the machine already happens to be running ypserv for the same
NIS  domain. Usually ypbind will bind to the NIS server on the same host,
although it doesn't have to be that way (if you have redundant NIS servers).

and 

Commands suggested are:

ypinit -m 	<- to start up NIS master

If they aren't then you'll need to rebuild the maps using your text  files:

cd /var/yp
make NOPUSH=1

and WWW pages referal:

http://www.ebsinc.com/solaris/network/nis.html

-----------------

Original Question:

Need help here, I recently sys-unconfig our NIS server. It is up and running
now but  I notice although I setup the our network with some help but I can
not
remember whether if I should also set NIS master as also client.

What I now notice is that since all NIS information are now available on
other computers on the network but it is not visible on the NIS server. I
assume that if start a
client up then NIS master should see all the DB.

Is there is/are a reason why I should setup my NIS master as client and if
so what?

May be I am talking rubbish but none of the users exists on NIS master and
what do I need to do so that I can see them 
again on NIS master. 

Can I have NIS master as client?
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From Ramiro.Santos at commerzbankib.com  Mon May  6 05:30:55 2002
From: Ramiro.Santos at commerzbankib.com (Santos, Ramiro)
Date: Mon, 6 May 2002 11:30:55 +0200
Subject: SUMMARY: DMA (IDE) on Solaris
Message-ID: <1D1B039860ABD41188C500508BAF2EB202483279@xmx9fraib.fra.ib.commerzbank.com>

Hi,
normaly people don4t use IDE based systems but unfortunately sun sells such
systems,
like Ultra 5, Ultra 10, Blade 100 and so on...
As a desktop system they are fine, but not performant.
Here is what I found about DMA on Solaris

> The ultra 5/10 models do not use udma or dma of any kind with regards to
> the ide implementation, PIO mode is used.  The blade workstations
> utilize the udma functionality.  The chipset for u5 -u10 is the older
> cmd640 without any bus mastering enhancements present.  Get a symbios
> 53c875spe card from ebay and go scsi for performance on these boxes.

> The CMD640 doesn't support UDMA but the CMD646B does. I've no idea
> what's inside of the currently shipping Ultra 10s, but four years ago
> I was explicitly told by the Sun engineer who ported and updated my x86
> ata driver to the SPARC to support and enable UDMA mode, that Sun was
> going to ship Ultra 10s with the CMD646 chip. The SPARC ata driver isn't
> as clean and robust as the x86 ata driver but I'm certain (unless some
> Sun engineer has recently patched the driver to disable it) that they
> both support UDMA mode on UDMA-capable hardware. Unfortunately, I'm not
> certain but it's possible that some of the U5/10 CDROM drives Sun sold
> do not support UDMA mode, and therefore do all CDROM I/O transfers in
> PIO mode (i.e., only the hard disk runs in UDMA mode). A CDROM drive
> running in PIO mode will cause a high CPU load (not a high I/O load).
> You could try checking the model number of your CDROM against the
> specs published by the drive vendor to see if it's UDMA-capable.


Many thanks to Richard Eisenman


Yours sincerely,

Ramiro Santos

ZIT P 7.41 Unix Server Implementation

_______________________________________________________

Contact:   Phone:     +49-(0)-69-136-43523
           Fax:       +49-(0)-69-136-47040
           E-Mail:    Ramiro.Santos at CommerzbankIB.com

Address:   Commerzbank AG
           Zentraler Servicebereich IT Production
           C/S Services Investment Banking - ZIT P 7.41
           Mainzer Landstr. 151 - DLZ 1, 06.52.0560
           D-60321 Frankfurt / Main
_______________________________________________________
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From plamadeleine at lightbridge.com  Mon May  6 11:17:06 2002
From: plamadeleine at lightbridge.com (Paul LaMadeleine)
Date: Mon, 06 May 2002 11:17:06 -0400
Subject: Initial Summary: OK, which secure ftpd do you like, Proftp,
  wu-ftp, ncftp or other
Message-ID: <5.1.0.14.0.20020506111057.01d58ec0@jupiter.credtech.com>

well,

         I've gotten a lot of responses and I have not had a chance to try 
them all.  I've initially gone with vcftp.  After downloading it and gcc to 
compile it, I got it installed and configured in around 15 minutes.  the 
only problem I'm having with it is I can't seem to make it to blind ftp 
(disable ls) without code changes.  However, the ls command is built in so 
I could just change the code not to do those calls.

         Below is the start of a real summary that I will post later.

         Thanks,

         Paul

ps - I did need to go into the code to change a couple of responses that 
seem inappriate on a production server.

>********
>   Here is my summary
>********
>
>  vsftpd:
>         website:        http://vsftpd.beasts.org/
>         cost   :
>        comments:
>
>   ncftpd:
>         website:        http://www.ncftpd.com
>         cost   :        $99 for < 50 connections, $199 for > 50, each per 
> server
>        comments:
>
>   pureftpd:
>         website:        http://www.pureftpd.org/
>         cost   :        free
>        comments:
>
>   pro-ftpd:
>         website:        http://www.proftpd.org/
>         cost   :        free
>        comments:
>
>   wu-ftpd:
>         website:        http://www.wu-ftpd.org/
>         cost   :        free
>        comments:
>
>   St. Louis ftpd:       can't find anything yet
>         website:
>         cost   :
>        comments:
>
>        surgeftp:
>         website:
>         cost   :
>        comments:
>
>   scp/sftp:
>     website:    http://www.openbsd.org
>         cost   :        free
>     comments:
>
>
>********
>   Responders
>********
>   Lars Hecking
>   Farouk Khawaja
>   Eduardo Sanchez M.
>   dana at dtn.com
>   William Yodlowsky
>   Andy_Bach
>   Frank Smith
>   Michael C. Ibarra
>   Scott Buecker
>   Hendrik Visage
>   Kruse, Jason K.





>Greetings,
>
>         I've been giving the task of setting up a secure ftp 
> server.  I've done a little research (which includes looking at the 
> archives) and now I would like your opinions.
>
>         I'd like to know how you thought of any of the ftp server's that 
> you may have used.  Please comment on installation/configuration, 
> maintinence, usablitity and security.  Even if you just rate them 1-5 
> would be cool.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From Robert.L.Harris at rdlg.net  Mon May  6 15:42:55 2002
From: Robert.L.Harris at rdlg.net (Robert L. Harris)
Date: Mon, 6 May 2002 13:42:55 -0600
Subject: SUMMARY: Console Card in a 280?
In-Reply-To: <3CD6B30F.F85861FF@mdh.si.edu>
References: <20020506095205.A29836@rdlg.net>
  <3CD6ABE8.E3F0D1AF@mdh.si.edu> <20020506104622.A31318@rdlg.net>
  <3CD6B30F.F85861FF@mdh.si.edu>
Message-ID: <20020506134255.A1194@rdlg.net>

In a sense,  BINGO.  It appears that someone had used my cable and
pulled out 1 pin.  I could get console text to my screen but nothing
from my keyboard would go to the console.  I changed cables and got a
good connection.  Over lunch took my cable adapter apart and found the
problem.



Thus spake Matt D. Harris (mdh at mdh.si.edu):

> Use a cat5 cable with the little Silver/Chrome DB25/RJ45 adapter that
> came with the 280R to console it.  I'm using that consoled directly to a
> Netra T1 (which has RJ45 serial ports) so console mine - I had the same
> issue you're having, actually.  Also, make sure there aren't any
> keyboards or such plugged in.  the OpenBoot input-device and
> output-device variables must be set to 'ttya', but I'm thinking your
> issue is a pin-out issue, not a software issue, as the pinouts for
> consoles have been a repeated thorn in my side for Sun gear.  Hope I've
> been even a little helpful!  :-)
> 
> "Robert L. Harris" wrote:
> > 
> > The problem is I can't get into the machine.  Console on TTYA won't take
> > input and the RSC has a password on it.
> > 
> > Thus spake Matt D. Harris (mdh at mdh.si.edu):
> > 
> > > The RSC board is something entirely different and freakish.  Once you
> > > have the box up and running check out the
> > > /usr/platform/*280R*/sbin/rscadm program, it lets you setup the RSC
> > > consoles.  If you open the box up, you'll notice that the RSC board has
> > > a 56k USR PCMCIA modem on it for the modem port.  :-)
> > > Basically it's a lot like LOM but much much more
> > > accessible/configurable.
> > >
> > > "Robert L. Harris" wrote:
> > > >
> > > > We have a 280 in to play with. Booted it up, connected to the ttyA
> > > > console as we normally do.  It gave some output and is waiting on a
> > > > login.  Not taking keyboard input.  At this point I notice the console
> > > > card, plug into it.  It's looking for a login/pass.
> > > >
> > > > Since we haven't set one I'm guessing it has a system default.  Anyone
> > > > have any idea what this is set to by default?
> > > >
> > > > :wq!
> > > > ---------------------------------------------------------------------------
> > > > Robert L. Harris                |  Micros~1 :
> > > > Senior System Engineer          |    For when quality, reliability
> > > >   at RnD Consulting             |      and security just aren't
> > > >                                 \_       that important!
> > > > DISCLAIMER:
> > > >       These are MY OPINIONS ALONE.  I speak for no-one else.
> > > > FYI:
> > > >  perl -e 'print $i=pack(c5,(41*2),sqrt(7056),(unpack(c,H)-2),oct(115),10);'
> > > > _______________________________________________
> > > > sunmanagers mailing list
> > > > sunmanagers at sunmanagers.org
> > > > http://www.sunmanagers.org/mailman/listinfo/sunmanagers



:wq!
---------------------------------------------------------------------------
Robert L. Harris                |  Micros~1 :  
Senior System Engineer          |    For when quality, reliability 
  at RnD Consulting             |      and security just aren't
                                \_       that important!
DISCLAIMER:
      These are MY OPINIONS ALONE.  I speak for no-one else.
FYI:
 perl -e 'print $i=pack(c5,(41*2),sqrt(7056),(unpack(c,H)-2),oct(115),10);'
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From atulgore at hotmail.com  Mon May  6 19:48:19 2002
From: atulgore at hotmail.com (Atul Gore)
Date: Mon, 06 May 2002 16:48:19 -0700
Subject: SUMMARY: Sun Certified Administrator
Message-ID: <F251dHQJYZ8cTYzfT5w0000a8e5@hotmail.com>

Gurus:

Pardon me for replying late. I have been suffering for one of worst
allergy seasons this year. Anyway.

I have passed both exams for Solaris8, 310-011 and 310-012 successfully.
I did not buy any exam crams neighter any braindumps. and I would like to
take this opportunity to thank the following gurus for their help:

Greg Hitchcock
Joon Martin Hansen
Bertrand_Hutin
Richard Skrabe
Tim Wort
Tony Tran
Amod Ozarkar
Mike
Doug Otto
Kevin Buterbaugh
Paul Foti

All the best to Mark Hargrave, Scott Pham, Steve Paige and Paul Wiggins
for your exams too.

Here is a list of pointers which I received:

http://www.brainbuzz.com
http://www.brainbench.com
http://brainbuzz.com
http://www.examslam.com
http://www.newriders.com/certification/
mailto:solaris8 at yahoogroups.com
http://groups.yahoo.com/solaris8/
http://www.coriolis.com/store/product.asp?sku=1281
http://www.learnsolaris.com
http://www.sunhelp.org
http://www.boson.com
http://www.cramsession.com
http://www.sunguru.com/

Manual: Solaris 8 Network Administrator Certification training guide, by
Rafeeq Rehman,
with a Cdrom.
New Riders, ISBN 1-57870-261-5

My two-cent addition:

http://bestdeals.tvheaven.com/solaris.htmhttp://www.solariscert.com

Thanks again All: -----------------------------------------------------
Best Regards... ... Atul

------------------------------------------------------------------------

Get your FREE download of MSN Explorer at http://explorer.msn.com.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From pnbadri at yahoo.com  Tue May  7 02:28:09 2002
From: pnbadri at yahoo.com (Badri narayanan)
Date: Mon, 6 May 2002 23:28:09 -0700 (PDT)
Subject: Summary: Suggestion required.
Message-ID: <20020507062809.92076.qmail@web12104.mail.yahoo.com>

Hai 

 I am thankful to the following list of people who had
spent their valuable time for me. 

1.Moti
2.Todd urie
3.Rick Waegner
4.Liam Baker
5.John England
6.Matthew Stier
7.Ashish Nanjiani
8.Sid Wilroy
9.David Levy.

     I am thankful to all others also who keep repling
my query.

   As everybody had quoted, I will try to increase the
swap space with the free space available with my
system. I had started the purchase and I am sure that
I may not be landing up with any problem, if at all
so, I have a complete listing of people backing up me
from the above list. I am thankful for the entire
group fo giving me such a valuable suggestions.

Thanks to one and all

Kindly forgive for my delay in this correspondence.

Bye
P.N.Badri narayanan

Note: forwarded message attached.
Yahoo! Health - your guide to health and wellness
http://health.yahoo.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From Callum.Hughes at sis.securicor.co.uk  Tue May  7 05:49:43 2002
From: Callum.Hughes at sis.securicor.co.uk (Callum Hughes)
Date: Tue, 7 May 2002 10:49:43 +0100
Subject: SUMMARY: Volume Manager and T3 
Message-ID: <E671455BAF0BFC43B8B31A6455348E7F512485@srvmeth10.sis.co.uk>

Hi everyone!

First of all, many thanks to the following people for their informative and
useful responses:

Rob Forsey
Darren Moulding
Tony Walsh
Eric Shafto
Wayne McCormick
Darren Dunham
Deborah Crocker
Charlie Giannetto
Christopher Ciborowski
Joe Serra
Doug Otto
Mailing List (???)
Vipin Sharma

I'll briefly summarize the two main points people made:

1) Volume Manager is only presented with LUN's and as a result, is not able
to see an individual disk failure in a RAID 5 configuration. My concern
however stemmed from the results of tests we performed on a RAID 0
configuration whereby the entire LUN did fail (as you would expect) when a
single disk failed. However, with the RAID 5 configuration, the T3 brings in
the hot spare and begins rebuilding the LUN automatically and completely
transparently as far as VM is concerned. The only concern voiced was that
there may be an impact on performance. 

2) The T3's can be configured as "partnered pairs" which enables you to use
the VM DMP feature. I don't know enough about T3's but this sounds very much
like multi-pathing on Clariion arrays. I might be completely wrong about
that one though so I'd recommend checking up on that if not sure! 

Just to see this all for myself, I pulled a disk from one of the T3's in a
RAID 5 configuration and yes, VM chugged away merrily whilst the T3 brought
in the hot spare and started building on that. VM was indeed completely
oblivious to the "failure" and no problems encountered. 

So I was quite a happy bunny really!

*Boing*

Callum A. Hughes
Unix Systems Engineer
e: callum.hughes at sis.securicor.co.uk



**********************************************************************
The information contained in this e-mail message is intended
only for the individuals named above.  If you are not the 
intended recipient, you should be aware that any 
dissemination, distribution, forwarding or other duplication 
of this communication is strictly prohibited.  The views 
expressed in this e-mail are those of the individual author 
and not necessarily those of 
Securicor Information Systems Limited.  
Prior to taking any action based upon this e-mail message 
you should seek appropriate confirmation of its authenticity.
If you have received this e-mail in error, please immediately 
notify the sender by using the e-mail reply facility.
**********************************************************************


_____________________________________________________________________
This message has been checked for all known viruses on behalf of Securicor Information Systems by the MessageLabs. For further information visit http://www.messagelabs.com or Email: mailsweeper.info at sis.securicor.co.uk
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From hps at hps.nu  Tue May  7 05:52:07 2002
From: hps at hps.nu (HP Schaechl)
Date: Tue, 07 May 2002 11:52:07 +0200
Subject: Summary: SunRay + USB barcode scanner
Message-ID: <5.1.0.14.0.20020507114545.00ab9270@mailserver.kippdata.de>

Hi!

My original question was whether someone got an
USB barcode scanner working on a SunRay.

Thanks to
Chris Hoogendyk <choogend at library.umass.edu>
krister bergman <krister.bergman at sun.com>

The consense seems to be that you'll need the very
latest SunRay firmware update with server software
v1.3 to make it work.
Otherwise you'll need to use something like a PS/2
to USB converter.
Sun has filed a bug (ID 4463574) which is closed
due to patch 111891-03 solving this issue.

Thanks!
Hans
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From McCaffityR at epg.lewis.army.mil  Tue May  7 10:56:27 2002
From: McCaffityR at epg.lewis.army.mil (McCaffity, Ray (Contractor))
Date: Tue, 7 May 2002 07:56:27 -0700
Subject: SUMMARY: SparcStation20 internal NIC devices?
Message-ID: <9669B05099E9D411B6E400B0D0AA476C5C37D1@epg.lewis.army.mil>

Thanks to the 2million who suggested "le0", this would have worked,
but I had to chage one other thing.  Go to the OK prompt and
setenv tpe-link-test? true,   reboot with a -r, bingo, it works.

Ray

================================================
>We have some "OLD" sparcstation20's that
>we were going through the AUI port to thin-net adapters
>for our network.  (I know, it's ancient, but it's what we had).
>So we finally upgraded to ethernet/cat5 this week.  All of these
>boxes have RJ-45 connectors on the back, but I can't seem to plumb them.
>These are the ports built-into the back on the box, not a card.
>
>I've tried, hme0, sr0, fe0, (I know they aren't qfe or ge)
>Does anyone know what the /dev/  NIC devices are?  How can I find out
>what they are?  Even though they are running Solaris 7, there is no
>prtdiag or prtconfig command.
>
>Ray
>_______________________________________________
>sunmanagers mailing list
>sunmanagers at sunmanagers.org
>http://www.sunmanagers.org/mailman/listinfo/sunmanagers
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From markhj01 at bahamut.thelair.dynamic-site.net  Sat May  4 20:58:50 2002
From: markhj01 at bahamut.thelair.dynamic-site.net (Dragon King)
Date: Sat, 4 May 2002 20:58:50 -0400 (EDT)
Subject: SUMMARY: Sparc 2 NVRAM
Message-ID: <Pine.GSO.4.10.10205042012140.9612-100000@bahamut>

Thanks to those who replied:
"Lumpkin, Buddy" <Buddy.Lumpkin at nordstrom.com>
Jacob Ritorto <jritorto at nut.net>
willief <willief at base-2.com>
system administration account <sysadmin at astro.su.se>

Recommended companies were:
Bell Microsystems (looked like they were in the UK)
Arrow Electronics
Base 2 Technologies

Also mentioned were Dallas Timekeeper chips. The faq at squirrel.com
mentions these fail the POST, but then can be booted fine after that. This
was confirmed by Jacob Ritorto.

I had wonderful dealings with Base 2, so special thanks to Buddy and
Willie for telling me about this company and furthermore to Willie for
getting the new chip on it's way to me.

To anyone else interested, Base 2 Technologies comes highly recommended by
me for customer service.

--

This message brought to you courtesy of the Dragon King.

http://thelair.dynamic-site.net
dragonking at carolina.rr.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From karyn at calarts.edu  Tue May  7 12:27:03 2002
From: karyn at calarts.edu (Karyn Williams)
Date: Tue, 07 May 2002 09:27:03 -0700
Subject: SUMMARY: Attachment sizes
Message-ID: <3.0.1.32.20020507092703.023a7e68@muse.calarts.edu>

Original post: I am planning to limit the size of attachments here and
thought it might be useful if I had an idea of what size other admins allow
in to their network. If I set our limit too high, it will just cause a lot
of bounces. So, would you mind letting me know what size you allow into
your network. Thanks.

Here are the votes:

100k		1
1MB		2
2MB		1
3MB		1
4MB		1	
4.5MB		1
5MB		3
10MB		2
15MB		1
30MB		1
60MB		1
none		2

As you can see from the responses, it is pretty much all over the place.
One person mentioned having one limit for internal mail and one for
outgoing mail, but I'm not sure how to do that at this time, so I will have
one limit for everything for now. I've decided to go with 10MB and see how
that flies. 


Thanks to:

Michael DeSimone
Stephen Smith
Todd Jensen
Michael Kriss
Lars Hecking
Kanchan Wadhwa
Tim Chipman
Sean Ryan
Allan West
Blaine Owens
Hendrick Visage
Moti Levi
Bob Rahe


-- 

Karyn Williams, CNE
Network Services Manager
karyn at calarts.edu
http://www.calarts.edu/network
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From Nelson.T.Caparroso at sbcdo.com  Tue May  7 15:02:18 2002
From: Nelson.T.Caparroso at sbcdo.com (Caparroso, Nelson T. (AAS))
Date: Tue, 7 May 2002 14:02:18 -0500
Subject: Summary: Serial Console & Graphics Console At The Same Time
Message-ID: <F469E17D26F4D51181620090278A0DBD709CCB@hqex01.aas.ameritech.com>

This "rare config" is indeed possible.


1. copy /usr/dt/config/Xservers to /etc/dt/config/Xservers

2. Change the last line in /etc/dt/config/Xserver
   from:     0   Local local_uid at console root /usr/openwin/bin/Xsun :0
-nobanner
   to:       0   Local local_uid at none root /usr/openwin/bin/Xsun :0
-nobanner

3. Then Change the following prom settings
     eeprom "output-device=ttya"
     eeprom "input-device ttya"

4. Also a need to change /etc/inittab entries for the new console to the
appropriate Terminal type "by defaut it is "-T sun" to something like "-T
vt100".

As we went enterprise-wide (which neccessitated Console Management) we also
needed to preserve existing investments on our LAN farm Sun KVM switch
system (we are able to keep our servers "headfull"). In such a setup, our
DataCenter folks continue to have access to a graphocal "console" while
sysadmins get to keep Console access...

Thanks to all that replied and apologies for this late Summary.

NELSON

>  -----Original Message-----
> Is it possible to have my console on Serial A while having Sun KB/Graphics
> access on headfull servers?
> 
> Thanks.
> 
> 
> NELSON
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From Nelson.T.Caparroso at sbcdo.com  Tue May  7 15:08:43 2002
From: Nelson.T.Caparroso at sbcdo.com (Caparroso, Nelson T. (AAS))
Date: Tue, 7 May 2002 14:08:43 -0500
Subject: SUMMARY: SE ToolKit for Solaris 8
Message-ID: <F469E17D26F4D51181620090278A0DBD709CCC@hqex01.aas.ameritech.com>

Thanks to all of you who've responded.

The consensus was that a majority of you are using the toolkit on Solaris 8
systems. It is just that the last time I've used SE was in 2.6 environments
and 2 years ago, the 64bit environments (Solaris 7/8) are not yet solidly
suppported.  Many of you also recommended a tool called ORCA
(www.orcaware.com) in conjunction with SE Toolkit

Thanks all..


>  -----Original Message-----
> List:
> 
> Any of you already using the SE toolkit for Solaris 8? How safe is it to
> use on prod environments? Just need it to collect basic stats ie. network,
> io, cpu states and use PERL/MRTG to do store and graph server stats...
> 
> 
> Thanks.
> 
> 
> NELSON
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From donunix at yahoo.com  Tue May  7 17:15:34 2002
From: donunix at yahoo.com (Don Jones)
Date: Tue, 7 May 2002 14:15:34 -0700 (PDT)
Subject: summary:nice/renice
Message-ID: <20020507211534.36238.qmail@web21304.mail.yahoo.com>

Thank you for all the answers:---- Original Message
>Hello all,>>I gave a user his own Ulra 2/Sol8 machine
so he

The answer is that the system IS using %100 of one
CPU, that's %50 of the 2 CPU!!.

-DJ


>can run his processes without competing with other
>users.  The ultra 2 refuses to give that user more
>than 50% of the CPU!.. even when I user renice, 
>he still cannot get more than 50% of the cpu.   The
>same process would take up to 90% on other systems.
>Could this problem be caused by a tweek to the
system?
>Is there a way to fix it so he can use the full
CPU>power?>>Thank you,>DJ
Yahoo! Health - your guide to health and wellness
http://health.yahoo.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From tmoran at shutterfly.com  Tue May  7 17:28:28 2002
From: tmoran at shutterfly.com (Tim Moran)
Date: Tue, 7 May 2002 14:28:28 -0700
Subject: Summary: Jumpstart boot server issues on second network
  interface .
Message-ID: <B5046A1066E6D311AC1A009027AA47B3030F2102@sfrc-exch.internal.shutterfly.com>

Thanks for the immediate feedback, got me up a going already!

The issue is that in.rarpd -a runs at boot time (/etc/rc3.d/S15nfs.server).
When the boot server was booted, the interface didn't exist (I brought it up
later).

Running in.rarpd -a again made the world happy.

Thanks again.
Tim

>  -----Original Message-----
> From: 	Tim Moran  
> Sent:	Tuesday, May 07, 2002 2:07 PM
> To:	'sunmanagers at sunmanagers.org'
> Subject:	Jumpstart boot server issues on second network interface.
> 
> I have a functioning boot server (that is, I've successfully jumpstarted
> servers using hme0) that I am attempting to use the second interface, for
> a different network.
> 
> I have brought up the interface and it is functioning well, I've added an
> entry in /etc/hosts and /etc/hostname.hme1, etc. The add_install_client
> accepts the arguments, and /etc/bootparams,  /etc/ethers and /etc/hosts
> look good, etc.
> 
> Snooping the second interface (snoop -d hme1) you can see the RARP go by,
> but the boot server never responds. I've removed and re-added the client
> multiple times, no response on hme0 either (that is, the boot server does
> not attempt to respond to the hme1 RARP on hme0).
> 
> Any thoughts on why the second interface would not work for a boot server?
> 
> Tim Moran
> tmoran at shutterfly.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From inzel at tunku.uady.mx  Tue May  7 17:56:01 2002
From: inzel at tunku.uady.mx (Israel Novelo)
Date: Tue, 07 May 2002 15:56:01 -0600
Subject: Summary /devices/pseudo 
Message-ID: <5.1.0.14.0.20020507155556.03d5f9e0@tunku.uady.mx>

Well thanks to all who help me :)

This summary

At 02:28 p.m. 07/05/02 -0400, you wrote:
>Each non-console login (telnet, rlogin, xterm, etc.) uses a pseudo-terminal
>(or pts.)  Each pts will be owned by the user who is logged in on it.  So
>yes, it is perfectly normal and acceptable to have some owned by users.
>
>For instance, type "who" and it will show you who is using which pts.


 > 1.- What is directory /devices/pseudo ? which is their propuse?
pseudo devices emulate other physical devices. In this case 'pts'
devices are described in the pts man page. They pretend to be a real
terminal connection and are used by telnet/rsh/ssh style programs that
connect to a 'terminal'.

 > 2.- Is normality that some users names are here?
Yes. /bin/login will usually chown the device to the user at login
time. If another user logs in, it should change to that user.
The 'grantpt' man page describes the process somewhat




>Regards,
>Charles
>
> > -----Original Message-----
> > From: Israel Novelo [mailto:inzel at tunku.uady.mx]
> > Sent: Tuesday, May 07, 2002 1:00 PM
> > To: sunmanagers at sunmanagers.org
> > Subject: A question...::)
> >
> >
> > Hi dear partners:)
> >
> > I hope you can help me...
> >
> >       I found this in /devices/pseudo,
> >
> > crw--w----   1 username1 tty       24, 11 May  7 11:32 pts at 0:11
> > crw--w----   1 root     tty       24, 12 May  7 11:40 pts at 0:12
> > crw-------   1 username2 tty       24, 13 May  7 11:46 pts at 0:13
> >
> >   and i want to know:
> >
> > 1.-  What is directory /devices/pseudo  ? which is their propuse?
> > 2.- Is normality that some users names are here?
> >
> > Thanks in advance... :)
> > _______________________________________________
> > sunmanagers mailing list
> > sunmanagers at sunmanagers.org
> > http://www.sunmanagers.org/mailman/listinfo/sunmanagers
> >


Thanks ALL :)
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From foster at dim.ucsd.edu  Tue May  7 18:19:54 2002
From: foster at dim.ucsd.edu (David Foster)
Date: Tue, 7 May 2002 15:19:54 -0700 (PDT)
Subject: [SUMMARY] Can't see large (>2GB) files from SGI systems
Message-ID: <200205072219.g47MJsw21572@dim.ucsd.edu>

I could not access large (> 2GB) files on filesystems exported from
our E420R Solaris 8 (10/01) systems (latest recommended patches),
from our SGI systems.

This turned to be an SGI issue, the diversity of knowledge on
this list never ceases to amaze me!

Disabling automount and enabling autofs on the SGIs fixed the problem,
which was that automountd was reverting to NFS version 2. We in fact
had the same problem on all Sun systems, I just didn't realize it
because in my testing I was mounting explicitly, bypassing the
automounter.

One person suggested patch 111197-02, but this replaces mountd and
fixes "4389722 nfs mount of a file > 2GB is impossible", an unrelated
issue.

I'll include the following very useful reply from Andy McVey
in its entirety:

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
The symptoms you describe are typical of an NFS version 2
mount rather than version 3.  Check the /etc/fstab for manual
mounts using vers=2 as an option.  The mount command reports
the version used:

bubble 43: mount | grep ghost
jaffa:/export/ghost on /home/ghost type nfs
(vers=3,rw,intr,soft,noquota,nosuid,dev=140001)

SGIs will try vers=2 if the server doesn't respond fast
enough.  This might happen if you're using Cisco switches and
you haven't enabled the fast port spanning tree algorithm on
the SGI's port.  You'll see a lot of "server not responding"
junk in the SYSLOG as the system boots up but eventually the
mounts will succeed as version 2 mounts.  Specify vers=3 in
the fstab/automap to fix this.  One other thing - make sure
you're using autofsd and not automountd:

bubblex 44: chkconfig | grep auto
        autofs               on
        automount            off

bubblex 45: ps -ef | grep auto
  mcveya      45574      45333  0 09:20:58 pts/0   0:00 grep
auto
    root      15788          1  0   Apr 04 ?       0:02
/usr/etc/autofsd -v -m 16

automountd is old, single threaded, unmaintained by SGI and
more likely to cause tears than any other SGI supplied daemon
(I worked in the support team for SGI Europe for several
years - that's a qualified statement!)
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -


In case enyone's interested, here's what I did to disable automount
and enable autofs (adjust config files for your site; assumes an NIS
source for "auto.direct"). 

Note that sometimes you have to run /usr/etc/autofs explicitly after a 
reboot to enable the new maps (this is supposed to be done automatically
on bootup).

#!/bin/csh -f
#---------------------------------------------------------------------------
# Disable automount and enable autofs on SGI systems
set hostname = `hostname`
echo "Disable automount and enable autofs for $hostname"
echo "Create /etc/auto_master"
cp /etc/auto_master /etc/auto_master.bak
cat >! /etc/auto_master << EOF
/home   auto.home       -rw,soft,intr
/-      auto.direct     -rw,soft,intr
EOF

echo "Create /etc/config/autofs.options"
cp /etc/config/autofs.options /etc/config/autofs.options.bak
set mounts = `ypcat -k auto.direct | grep $hostname | awk '{print $1}'`
set arg = "-v -m 16"
if ( $#mounts > 0 ) then
        foreach m ( $mounts )
                set arg = "$arg $m -null"
        end
endif
echo $arg >! /etc/config/autofs.options

echo "Disable automount and enable autofs"
/etc/chkconfig -f automount off
/etc/chkconfig -f autofs on

echo "Reboot system for changes to take effect."
exit 0



Thanks to:

Matt D. Harris
Andy McVey
Osama Ahmed
Weigert, Daniel
Robert Konecny

> 
> Platform: SunOS cressida 5.8 Generic_108528-12 sun4u sparc SUNW,Ultra-80
> 
> Problem: 
> 
>    On two E420R systems with recent recommended patches applied and
>    at Solaris 8 (10/01), we cannot see large (> 2GB) files on their
>    NFS mounts from our SGI systems running IRIX 6.5.14:
>    
>       $ ls -l /auto/keck1f
>       ls: /auto/keck1f/data3G.tif: File too large
> 
>    Other file utilities fail, as do basic C system calls.
>    
>    The NFS mounts are both T3 and A5100 disk arrays (only A5100's under
>    Veritas VM control), so this isn't related to VxVM. All disk
>    arrays are attached via fibre.
>    
>    I don't think this is an IRIX problem because we can see large
>    files on shares from all of our other Sun systems (U80, U60, E450).
>    What's even stranger, we can also see large files on the E420Rs'
>    system disk.
>    
>    The only difference with these two systems is that they were
>    set up for us by Sun SE personnel. The recommended patch cluster
>    installed on them is more recent than some of our other systems, so 
>    I'm stumped.
>    
> Has anyone else seen this behavior before?
> 
> Dave Foster


   << All opinions expressed are mine, not the University's >>

  =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
   David Foster    National Center for Microscopy and Imaging Research
    Programmer/Analyst     University of California, San Diego
    dfoster at ucsd.edu       Department of Neuroscience, Mail 0608
    (858) 534-7968         http://ncmir.ucsd.edu/
  =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=

   "The reasonable man adapts himself to the world; the unreasonable one
   persists in trying to adapt the world to himself.  Therefore, all progress
   depends on the unreasonable."   -- George Bernard Shaw
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From beechanji at hotmail.com  Tue May  7 22:56:01 2002
From: beechanji at hotmail.com (Johny Smith)
Date: Wed, 08 May 2002 02:56:01 +0000
Subject: SUMMARY : Sendmail vs other smtp
Message-ID: <F128ncJfKHmj9n1xxWn00001125@hotmail.com>

Here are expert ideas I got regarding this below question

QUESTION
========
I am creating new mail server/set up for my client (ISP) n wondering which
is the best  for mailserver either sendmail or postifx or  qmail or any
other n also suggest pop3 server , according to ur experience specially in
ISP.


ANS
===

From unixbeginer at hotmail.com  Tue May  7 23:38:57 2002
From: unixbeginer at hotmail.com (beginer unix)
Date: Wed, 08 May 2002 03:38:57 +0000
Subject: SUMMARY: SUN Trunking
Message-ID: <F116tPBjxPGsKsJbB8q00007c2f@hotmail.com>

Hi all,

Thanks to:
Brian Sherwood
Steve Mickeler
Suhas Bhide
Steve Camp

IP Multipathing is what I was looking for

Thanks

Sunil
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From makhlooq at batelco.com.bh  Wed May  8 00:41:51 2002
From: makhlooq at batelco.com.bh (Makhlooq)
Date: Wed, 8 May 2002 07:41:51 +0300
Subject: Summary:New H/D
Message-ID: <007e01c1f64a$ad74cb90$8e70bcc1@DELL>

Thanks to all who response to me I managed to solve the problem, which was
only the save command was not commited after formating and labeling etc ....
Thanks again

Hussain Makhlooq
STO System Management ( NI21 )
Office :+973-883063
Mobile : +973-9685646
Fax    :+973-9103063
74/`7.8.74/`7.8 ><((((?>
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From whoracle at dialup.ptt.ru  Wed May  8 01:24:52 2002
From: whoracle at dialup.ptt.ru (Grigory Nikonov)
Date: Wed, 08 May 2002 09:24:52 +0400
Subject: SUMMARY: A1000 write cache battery problem
Message-ID: <3CD8B6A4.D2A060D4@dialup.ptt.ru>

Hi everybody.

thanks to all who replied:

Krister Bergman
Suresh Kakkar
Mike Salehi
Mark Montague
Steve Wills
Matthew Sams

the correct solution was to wait some more time - 30 hours in fact - for
the battery to charge.
everything works now. thanks again.

Grigory Nikonov wrote:
> 
> Hi everybody.
> 
> I've got a problem with a new battery for A1000 storage array.
> 
> I've installed the battery, run raidutil -c c1t0d0 (d1,d2) -R.
> 
> raidutil -B shows that battery age is between 0 and 90 days.
> 
> but healthck keeps showing Battery Alert. I installed the battery more
> than 12 hours ago,
> so I think it should've been charged by now...
> 
> thanks.



_________________________________________________
Grigory Nikonov
Developer & System Administrator
Renaissance Insurance, Moscow, Russia
Tel:         +7 095 725 1059
Fax:         +7 095 967 3535
Web-site:    www.renins.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From stephen.t.hughes at au.unisys.com  Wed May  8 01:29:35 2002
From: stephen.t.hughes at au.unisys.com (Hughes, Stephen T.)
Date: Wed, 8 May 2002 15:29:35 +1000
Subject: SUMMARY: Multiple IP addressing to same device - Solaris 8
Message-ID: <FD6779273196D211932F00E02915B5280446A8BB@au-bne-exch-1.au.unisys.com>

Thanks to those that responded so quickly.

We needed to 'plumb' each virtual device.

Cheers
Stephen 

-----Original Message-----
From: Hughes, Stephen T. [mailto:stephen.t.hughes at au.unisys.com]
Sent: Wednesday, 8 May 2002 15:02
To: 'sunmanagers at sunmanagers.org'
Subject: Multiple IP addressing to same device - Solaris 8 


Hi,

I am in need of assistance in getting multiple IP addresses assigned to a
single network card. We have just discovered that Solaris 8 does not allow
this facility which is available under 2.6 and below. 

If you can provide any help in how to acheive this we would appreciate it.

Regards
Stephen Hughes
Unix Systems Admin
Unisys - TAFE Queensland Project
South Bribane
Queensland
Australia
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From jonas.blaberg at cellnetwork.com  Wed May  8 03:27:42 2002
From: jonas.blaberg at cellnetwork.com (=?iso-8859-1?Q?Jonas_Bl=E5berg?=)
Date: Wed, 8 May 2002 09:27:42 +0200
Subject: SUMMARY: HW error on my E250?
Message-ID: <E7491F05BCBA0B41B40E23AF1039FE1E8845C0@exgbgmb01.europe.cellnetwork.com>

Hello!

I got this message from lots of people: "Use prtdiag -v"

The strange thing is that the output from prtdiag -v differ from the actual
LEDs on the front panel. POWER SUPPLY ERROR says OFF but it is still yellow!

Also a manual
(http://sundocs.princeton.edu:8888/ab2/coll.240.2/UGJAVELIN/@Ab2PageView/idma
tch(Z400037430C1)#Z400037430C1?) said that the General Fault LED should be
yellow if any other fault has been detected. It is not yellow!

I also got some replies with hints of what to do if there is an inconsistancy
between LEDs and prtdiag -v output:

	The prtdiag command (/usr/platform/sun4u/sbin/prtdiag) should show you which
	component is faulty.  I have had this happen on an E250 when I interrupted
the
	boot process, when there was not actually anything faulty.  The light stayed
	on, but there was no fault.  The solution I used to get it to go out was to
	physically power it off, and then start it up again.  This may not be an
	option, but if you can reboot the machine would be a useful test if prtdiag
	doesn't show an error.

	Run `/usr/platform/sun4u/sbin/prtdiag -v |more`
	This will identify any hardware problems you may have, with very few
	exceptions.  If the wrench light remains orange and yet prtdiag does not
	report any problems, try running SunVTS or somesuch.  If nothing comes
	up, call in Sun support.


Many thanks to:

Graham Wood [rauxon at dragonhold.org]
Tommy.Fallsen at kongsberg.com
Mike McShane [mcshane at cs.umn.edu]
Matt D. Harris [mdh at mdh.si.edu]
Bertrand_Hutin at notes.amdahl.com
Steve Wills [steve at stevenwills.com]
Bob Rahe [bob at dtcc.edu]
Matthew SAMS [maclean at cs.mcgill.ca]
VGunville at Lumenos.com
Christophe Dupre [duprec at scorec.rpi.edu]

/jonas


>  -----Ursprungligt meddelande-----
> Fren: 	Jonas Bleberg
> Skickat:	den 7 maj 2002 08:32
> Till:	'sunmanagers at sunmanagers.org'
> Dmne:	HW error on my E250?
>
> Hello!
>
> I have an E250 where the front-end lamp marked with a "flash" symbol has
gone orange. Both power supplies are showing two green lights. I have not
found anything in the messages log (and syslog should be configured to put
everything interesting there). Nothing is behaving strange in the machine. I
installed an extra NIC two weeks ago, but it is working as it should.
>
> Which units are showing problems through the "flash" symbol on the front?
>
> Is there any way to find out which unit is erroneous?
>
> /jonas
>
>
>
> Jonas Bleberg
>   Cell Network Sverige AB
>   Kruthusgatan 17,6
>   S-411 04 Gvteborg
> jonas.blaberg at cellnetwork.com
> @office: +46-(0)31-739 84 54       <----- Number has changed!
> not at office: +46-(0)709-95 00 68
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From nc at ncartron.com  Wed May  8 05:18:22 2002
From: nc at ncartron.com (Nicolas Cartron)
Date: Wed, 8 May 2002 09:18:22 -0000 (GMT)
Subject: SUMMARY : Mounting a CDROM
Message-ID: <58253.213.41.92.76.1020849502.squirrel@www.42bsd.org>

Thanks to all managers who answered me very quickly :

Pierre Zimmermann
Stefaan Margot
Callum Hughes
Adam Kirby

(the right command for me was
  mount -F hsfs -r /dev/dsk/c0t6d0s0 /cdrom

I've got a last question : how can i know the device's name
(/dev/dsk/c0...) ? I looked at dmesg but didn't find,
neither in /var/adm/messages*

--
Nicolas CARTRON
Administrateur Unix & Linux
GSM : +33.6.63.66.11.77
--
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From nc at ncartron.com  Wed May  8 06:48:06 2002
From: nc at ncartron.com (Nicolas Cartron)
Date: Wed, 8 May 2002 10:48:06 -0000 (GMT)
Subject: SUMMARY 2 : Mounting a CDROM]
Message-ID: <62983.213.41.92.76.1020854886.squirrel@www.42bsd.org>

> Question was :I've got a last question : how can i know the device's name
> (/dev/dsk/c0...) ? I looked at dmesg but didn't find,
> neither in /var/adm/messages*

Answer is : iostat -En

Thanks again.


--
Nicolas CARTRON
Administrateur Unix & Linux
GSM : +33.6.63.66.11.77
--
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From Stan.Pietkiewicz at statcan.ca  Wed May  8 09:16:36 2002
From: Stan.Pietkiewicz at statcan.ca (Stan.Pietkiewicz at statcan.ca)
Date: Wed, 8 May 2002 09:16:36 -0400
Subject: Tentative Summary: Bootable Tape 
Message-ID: <CAA8A9B10AE1D511995A0002A53F974BCACE91@msxa4.statcan.ca>

I had a quick look at this document, and it looks like it might be helpful.
I haven't tried it yet, and I'm not sure if it will write a bootable tape.

http://www.sun.com/solutions/blueprints/1101/webstart.pdf


The other responses from the SunManagers mailing list ranged from: "Not
possible" to "Let me know when you find out".

Stan Pietkiewicz
Stan.Pietkiewicz at statcan.ca
Informatics Technology Services Division - Statistics Canada
It may be statistically possible that my opinion is the same
as someone else's - but it is still my opinion!


-----Original Message-----
From: Pietkiewicz, Stan - ITSD/DSTI 
Sent: May 7, 2002 11:53 AM
To: 'sunmanagers at sunmanagers.org'
Subject: Bootable Tape 


I was asked by one of my co-workers about a bootable tape for Solaris.... Is
there a way to write a bootable tape image of a root drive / filesystem for
recovery purposes?

Thanks.....

Stan Pietkiewicz
Stan.Pietkiewicz at statcan.ca
Informatics Technology Services Division - Statistics Canada
It may be statistically possible that my opinion is the same
as someone else's - but it is still my opinion!
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From jeffw at pbs.com  Wed May  8 11:28:30 2002
From: jeffw at pbs.com (Jeff Wellens)
Date: Wed, 8 May 2002 10:28:30 -0500
Subject: SUMMARY- Disksuite problem.
Message-ID: <NEEBKGEBEINHGKGIGFLCCEKACIAA.jeffw@pbs.com>

First of all, Thanks to everyone who replied and took time to help me with
this.
Huge thanks to Steve Camp!!

After further review, I found that at some point external disk were
connected up wrong.
The format command reveled that I had a c2t3d0s6 and a c3t3d0s6 and NO
c1t3d0s6.

To resolve, I ran the following commands:

1. ensured I had a good backup of this filesystem
2. metareplace d80 c1t3d0s6 c3t3d0s6  (wait until state was ok)
3. metareplace -e d80 c2t2d0s6
-------------------------------------------------


 -----Original Message-----
From: Jeff Wellens [mailto:jeffw at pbs.com]
Sent: May 07, 2002 8:58 AM
To: sunmanagers at sunmanagers.org
Subject: Disksuite problem.


Hello All -

I have an issue with disksuite where both sub-mirrors are in a "Needs
Maintenance" state.
The funny thing is that it may have been this way for a while as the mounted
filesystem for
this device is working just fine.   Both drives appears fine as well.

What is the best way for me to go about resolving this and not lose any data
on this filesystem?


Here is the output of metastat for this metadevice:

d80: Mirror
    Submirror 0: d81
      State: Needs maintenance
    Submirror 1: d82
      State: Needs maintenance
    Pass: 1
    Read option: roundrobin (default)
    Write option: parallel (default)
    Size: 35358848 blocks

d81: Submirror of d80
    State: Needs maintenance
    Invoke: metareplace d80 c1t3d0s6 <new device>
    Size: 35358848 blocks
    Stripe 0:
        Device              Start Block  Dbase State        Hot Spare
        [Jeff Wellens]  0     No    Maintenance


d82: Submirror of d80
    State: Needs maintenance
    Invoke: after replacing "Maintenance" components:
           metareplace d80 c2t3d0s6 <new device>
    Size: 35358848 blocks
    Stripe 0:
        Device              Start Block  Dbase State        Hot Spare
        c2t3d0s6                   0     No    Last Erred


Thanks!


Jeff Wellens
Technical Support Analyst
jeffw at pbs.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From Nelson.T.Caparroso at sbcdo.com  Wed May  8 14:32:16 2002
From: Nelson.T.Caparroso at sbcdo.com (Caparroso, Nelson T. (AAS))
Date: Wed, 8 May 2002 13:32:16 -0500
Subject: SUMMARY: Any danger booting 167Mhz U2 Systems 64-bit Solaris 8?
Message-ID: <F469E17D26F4D51181620090278A0DBD709CD8@hqex01.aas.ameritech.com>

Mnay of you are saying that apart from the "theorethical" possibility of a
code stalling the CPU, it should be perfectly okay to run the 167Mhz based
UltraSparc II based systems (mine are Ultra Enterprise 2's) in 64-bit mode.
To those that are'not aware how this is done, it is actually displayed at
startup - to read the man pages on boot (4m) and is as simple as creating
the boot.conf file under the /platform/<platform_name> directory with the
following:

ALLOW_64BIT_KERNEL_ON_UltraSPARC_1_CPU=true

But as far as the benifits in doing so, it all depends on what your needs
are. Mine was just the safety and stability part of it in addition to we are
actually finding the 64-bit drivers and modules to be more robust than the
32-bit ones...

Thanks all.


NELSON



>  -----Original Message-----
> I've a couple of U2's which I've upgraded to Solaris 8... I added a
> boot.conf to force it to boot 64-bit always. Is there any danger to
> forcing these systems to run the 64-bit kernel (which basically is 32/64
> anyways..)?
> 
> Thanks.
> 
> 
> NELSON
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From Alex.Wilkinson at dsto.defence.gov.au  Wed May  8 22:11:34 2002
From: Alex.Wilkinson at dsto.defence.gov.au (Wilkinson,Alex)
Date: Thu, 9 May 2002 11:41:34 +0930 (CST)
Subject: SUMMARY:  Increasing number of processes ?
Message-ID: <20020509113533.X92839-100000@squirm.dsto.defence.gov.au>

Thanks to
---------
ed at the7thbeer.com
scotrn <scotrn at cox.net>
amit mahajan <amahajan at santelnet.com>
Mike's List <mikelist at sky.net>
Wolf Schaefer <schaefer4 at llnl.gov>



Original Question
-----------------

I have a problem where one user of a group I belong to spawned a lot of processes
without killing them. When I tried to log in and vfork reported a failure
(No process available). I logged in as root and rebooted the machine.

Is there any way to increase the number of processes (PID)
available to a group of users?

System is:

Sun Os 5.5 Generic_103093-27 sun4m sparc, SPARC, CPU-5V


Answers
-------

You probably want to look at three tunables.  In short order.

max_nprocs:  derived as 10+(16 x maxusers)
maxusers: see above.  Affects many other tunables, however.
reserved_procs: numbers of reserved processes that can run in the process
table, only as UID 0.  Defaults to 5, I ususally run the box, ps -ef and
grep for UID 0/root processes, add 10 and change in /etc/system in the
event of just such an issue you experience. :)

.....

The maxusers kernel parameter is the one most often tuned. By default, it is
set to the number of Mb of physical memory or 1024, whichever is lower. It
cannot be set higher than 2048.
Several kernel parameters are set when maxusers is set unless otherwise
overridden by the /etc/system file. Some of these formulas differ between
different versions of Solaris:

max_nprocs: Number of processes = 10 + (16 x maxusers)

ufs_ninode: Inode cache size = (17xmaxusers)+90 (Solaris 2.5.1) or
4x(maxusers + max_nprocs)+320 (Solaris 2.6-8).

ncsize: Name lookup cache size = (17xmaxusers)+90 (Solaris 2.5.1) or
4x(maxusers + max_nprocs)+320 (Solaris 2.6-8).

ndquot: Quota table size = (maxusers x 10) + max_nprocs

maxuproc: User process limit = max_nprocs - 5

.....

I think you can edit /etc/system and have the following line:
set pt_cnt=256
set maxusers=64

Not sure if you need both or not, but you can try it and see...

/etc/system = the kernel, so you'll have to reboot, might want
to search the www.sunmanagers.org archive first (not google.com)
and docs.sun.com and sunsolve.sun.com

....


Add the following line to your /etc/system file and reboot with
option -r.  From the ok prompt boot -r, from the UNIX command
line reboot -- -r.

set pt_cnt=127


Wolf


Thanks

 - Alex
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From Alex.Wilkinson at dsto.defence.gov.au  Wed May  8 22:54:46 2002
From: Alex.Wilkinson at dsto.defence.gov.au (Wilkinson,Alex)
Date: Thu, 9 May 2002 12:24:46 +0930 (CST)
Subject: SUMMARY: PROM Doc
Message-ID: <20020509121910.N92941-100000@squirm.dsto.defence.gov.au>

Thanks to:
---------
Mike's List <mikelist at sky.net>
Gabel Martin <Martin.Gabel at ffm3.siemens.de>
Anna Berger <berger2416 at yahoo.com>
Matthew SAMS <maclean at cs.mcgill.ca>
Moti <moti at flncs.com>
David Foster <foster at dim.ucsd.edu>
John England <john_w_england at yahoo.com>


Original Question:
-----------------
Just a quick query. Can anyone point me to a good paper on
using the PROM ?


Answers
-------

sun aswerbook
sunsolve.sun.com
http://docs.sun.com/ab2/coll.216.2/@Ab2CollView?Ab2Lang=C&Ab2Enc=iso-8859-1
http://www.squirrel.com/squirrel/sun-nvram-hostid.faq.html

Thanks to all who replyed.

 - Alex
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From mtdelavega at pse.org.ph  Thu May  9 05:33:38 2002
From: mtdelavega at pse.org.ph (Maria Aurora VT de la Vega)
Date: Thu, 09 May 2002 17:33:38 +0800
Subject: Summary: CPU's
References: <3CDA22F1.F0232008@pse.org.ph>
Message-ID: <3CDA4272.C505F920@pse.org.ph>

Thanks everyone!

and the winner is .... psrinfo!

=)

Maria Aurora VT de la Vega wrote:

> Hello Sunmanagers...
>
> quick question.
> What's the command to find out the # of CPU's on our Sun E450 machine?
>
> Thanks.
> =)
> --
> Maria Aurora VT de la Vega (OCP)
> Database Specialist
> Philippine Stock Exchange, Inc.
> _______________________________________________
> sunmanagers mailing list
> sunmanagers at sunmanagers.org
> http://www.sunmanagers.org/mailman/listinfo/sunmanagers

--
Maria Aurora VT de la Vega (OCP)
Database Specialist
Philippine Stock Exchange, Inc.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From Callum.Hughes at sis.securicor.co.uk  Thu May  9 09:13:37 2002
From: Callum.Hughes at sis.securicor.co.uk (Callum Hughes)
Date: Thu, 9 May 2002 14:13:37 +0100
Subject: SUMMARY: FSCK error for non existant file system
Message-ID: <E671455BAF0BFC43B8B31A6455348E7F5124D2@srvmeth10.sis.co.uk>

Thanks to all who replied:

Justin Stringfellow 
Tony Walsh
Stan Pietkiewicz
Mike Demarco
John Riddoch
Rob Forsey
Phil Rainey
Mike Scott
Steve Sandau
Dave Mitchell
Darren Moulding

Most suggestions were to re-examine the /etc/vfstab to make sure that the
device and the device to fsck were consistent. However, top award for this
one goes to Mr Tony Walsh for his SUPERB instructions on eradicating shadow
devices from the PROM, posted, for your viewing pleasure below!

The first experiment is to shut the machine down to the OBP and issue the
following commands at the OBP prompt.
OK> setenv auto-boot? false
OK> setenv diag-level max	<-- This may not work depending on you OBP
version
OK> setenv diag-switch? true
OK> power-off
Now use the soft power button on the front of the U10 or use the power
switch to turn it on. You will now have to wait several minutes while the
extended diags complete. When they do finish, issue the following commands
and reboot.
OK> setenv diag-level min
OK> setenv diag-switch? false
OK> reset-all
Wait for the OBP prompt to return and issue the following.
OK> setenv auto-boot? true	<-- You could leave this out if you want.
OK> boot
Now wait for the offending message. If it does not reoccur, then this
procedure has just cleared a shadow device from the OBP device tree which
was being seen by the OS and interpreted as a drive needing attention. Also
if it does not reoccur, this is a sign that the OBP level you have is a
little old and is probably full of bugs that are capable of presenting these
'shadow devices' (so upgrade the OBP).

A possible reason for this device difference, is that when you replaced the
second drive you may have set (or left) jumpers on the drive that indicate a
Master/Slave IDE setup instead of the 'Cable Select' setup favoured by Sun.
You may also have changed the cabling around so that the new drive is on the
second IDE channel and not the primary channel.

If none of the above works, try the following as a second experiment.

First move aside or remove the file called /etc/path_to_inst and
/etc/path_to_inst.old. 
Then issue 'reboot -- -svar' or 'boot -svar' from the OK prompt and respond
to the questions as required (Mostly just hit enter). When asked if you want
to rebuild the path_to_inst file, reply yes and carry on.
This process should rebuild the path_to_inst file from scratch and rebuild
the OS device tree.

How cool is that??? Well cool! Thanks again to all!


Callum A. Hughes
Unix Systems Engineer
Securicor Information Systems
t: 01249 - 665 -396
e: callum.hughes at sis.securicor.co.uk



**********************************************************************
The information contained in this e-mail message is intended
only for the individuals named above.  If you are not the 
intended recipient, you should be aware that any 
dissemination, distribution, forwarding or other duplication 
of this communication is strictly prohibited.  The views 
expressed in this e-mail are those of the individual author 
and not necessarily those of 
Securicor Information Systems Limited.  
Prior to taking any action based upon this e-mail message 
you should seek appropriate confirmation of its authenticity.
If you have received this e-mail in error, please immediately 
notify the sender by using the e-mail reply facility.
**********************************************************************


_____________________________________________________________________
This message has been checked for all known viruses on behalf of Securicor Information Systems by the MessageLabs. For further information visit http://www.messagelabs.com or Email: mailsweeper.info at sis.securicor.co.uk
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From cbar44 at tsg.cbot.com  Thu May  9 13:25:52 2002
From: cbar44 at tsg.cbot.com (Christopher L. Barnard)
Date: Thu, 09 May 2002 12:25:52 -0500
Subject: SUMMARY: ssh question 
In-Reply-To: Message from "Christopher L. Barnard" <cbar44@tsg.cbot.com>
     of "Thu, 09 May 2002 10:25:41 CDT."
  <200205091525.KAA24987@shalmaneser.tsg.cbot.com> 
Message-ID: <200205091725.MAA25453@shalmaneser.tsg.cbot.com>

I asked:

> I have a question concerning OpenSsh 3.0.2 and 3.1 (p1) on Solaris 7 and
> Solaris 8 servers.  I am able to connect from machine A to machine B,
> but not from machine B back to A.  This occurs on only a handful of
> machines, and I have not detected any pattern as for as OS level or
> OpenSsh version.  On all machines the /etc/ssh_config file these entries
> are changed from the default:
> 
> ForwardAgent no
> ForwardX11 no
> UsePrivilegedPort yes
> StrictHostKeyChecking no
> 
> and the etc/sshd_config these entries are changed from the default:
>
> IgnoreRhosts no
> X11Forwarding yes
> X11DisplayOffset 10
> PrintMotd no
> 
> the error message when I try to connect is
> 
> ssh_exchange_identification: Connection closed by remote host
> 
> any ideas what else to check? TIA and I will summarize of course.

The answer:

<dolt slap>
the /etc/hosts.allow file of course.  These new boxes are on a new network and I had forgotten to update my hosts.allow file.  

Thanks to:

the millions of people who graciously pointed out the obvious to me.

+-----------------------------------------------------------------------+
| Christopher L. Barnard         O     When I was a boy I was told that |
| cbarnard at tsg.cbot.com         / \    anybody could become president.  |
| (312) 347-4901               O---O   Now I'm beginning to believe it. |
| http://www.cs.uchicago.edu/~cbarnard                --Clarence Darrow |
+----------PGP public key available via finger or PGP keyserver---------+
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From jrodriguez at intellinet-tech.com  Thu May  9 14:15:29 2002
From: jrodriguez at intellinet-tech.com (Jeremy Rodriguez)
Date: Thu, 9 May 2002 14:15:29 -0400
Subject: SUMMARY: firewall
Message-ID: <HMEIJMEKLHGNJPCDCEBOOECMDAAA.jrodriguez@intellinet-tech.com>

I asked:
>I am running Solaris 7 sparc, in 32-bit mode and would like a freeware
firewall.
>I looked at BigAdmin and there was a mention of PhoneBoy's firewall, just
wondering has anybody used it?
>What is the standard for setting up a firewall on Solaris 7? links,
directions, and a pointer some >>  > freeware would be a great help.

Responses:
looks like hands down everyone choose ipfilter.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From gio at uslink.net  Thu May  9 15:18:30 2002
From: gio at uslink.net (Giovanni Navarrette)
Date: Thu, 9 May 2002 14:18:30 -0500 (CDT)
Subject: SUMMARY: Laptop console for Solaris SPARC server?
Message-ID: <Pine.GSO.4.05.10205091354590.6892-100000@netapp>

Hey everyone:

Thanks for the swarm of responses! I love this list :D

My issue was basically what EVERYONE said: null modem adapter. I got
output (but no input) with the Cisco 678 console cable I had (its
different from a normal Cisco console cable), so I tried this and it
worked:

sun-box -> null modem adapter -> straight through rj-45 ethernet ->
rj45-db25 adapter -> db25-to-db9 -> pc running secureCRT.

I'm sure I can simplify it, but for now I'm jumping up and down with glee
as it works, so now I can try to simplify it and make my cable with less
parts. Thanks to EVERYONE who replied, y'all are great!!!

Have an excellent day!

-------------------------------------
Giovanni Navarrette
USLink Internet Systems Administrator
Email: gio at uslink.net
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From kartano at ppllc.com  Thu May  9 17:20:49 2002
From: kartano at ppllc.com (Tom Kartanowicz)
Date: Thu, 9 May 2002 17:20:49 -0400
Subject: SUMMARY: /var file removal question
Message-ID: <1E8A366292968B4EB4ED5DAE2682B5E56E216F@HAMILTON>

Many thanks to Somesh Nagthan, Hemant, Elizabeth Lee, Steve Wills ,Ramini
Gabriel and Bryan L. Moore.
for quick and informative responses.

The short answer is : Don't remove those files unless you want to loose
control of system software.

Thanks!

Replies follow:

All the package administration tools (pkgrm, pkgadd, ecc)
depends on the contents of that directory.
Removing it means to loose control of the system software... 

Removing them is probably a bad idea if you plan on administering the box
cleanly in the future, installing patches, etc. I'm guessing you need disk
space in /var? Try moving the sadm dir to somewhere with more space and
making
a symlink to it.

better leave /var/sadm/pkg alone -- in fact, leave /var/sadm alone.

Do not touch /var/sadm directory.


the package administration commands use this directory.....so i would
strictly advise you to NOT mess with it at all.

I believe when you remove the record of it being installed, if you ever do a
pkgrm on the package, it comes back saying no package of that name
exists....
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From jseparov at uecomm.com.au  Thu May  9 18:51:26 2002
From: jseparov at uecomm.com.au (Separovic, Jason)
Date: Fri, 10 May 2002 08:51:26 +1000
Subject: SUMMARY: Detecting a power supply failure on an E220R.
Message-ID: <2B5FDC783FD3D511AB07000629A847A33DE64B@MAILPRD03>

> It seems that the spanner LED on the front of the server is the only way
> of knowing that you have a faulty PSU or a loss of power to a PSU. 
> 
> Thanks to all who replied.
> 
> 
> 
> 	-----Original Message-----
> 	From:	Separovic, Jason 
> 	Sent:	Thursday, May 09, 2002 1:50 PM
> 	To:	'sunmanagers at sunmanagers.org'
> 	Subject:	Detecting a power supply failure on an E220R.
> (Further Info)
> 
> 	Hi again,
> 
> 	Check this out from docs.sun.com "Sun Enterprise 220R Server Owner's
> Guide":
> 	================================================
> 	The system provides fault LEDs in the following places:
> *	Front panel
> *	Keyboard
> *	Power supplies
> 	Error messages are logged in the /var/adm/messages file and are also
> displayed on the system console by the diagnostic tools.
> 	================================================
> 	Is this true? Will a faulty PSU will be noted in /var/adm/messages?
> 	If anyone has seen this please let me know. Or even better please
> send me a copy of the error message.
> 	Or if anyone currently has a faulty PSU ... well you could test out
> Sun's documentation if you wanted to.
> 	Please send email and I will summarise to the list.
> 
> 
> 
> 	From the replies and testing so far:
> 		- "prtdiag -v" does not show PSU status
> 		- loss of power feed to PSU does not show up in
> "/var/adm/messages"
> 		- "SPANNER" LED is the only indication.
> 
> 
> 	Original Message:
> 
> 	I have an Enterprise E220R Workgroup Server with redundant power
> supplies.
> 	Does anyone know how to detect the following scenarios via the OS:
> 		- Loss of power feed to one of the supplies
> 		- Faulty power supply
> 	Please send email and I will summarise to the list.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From Dennis_Martens at health.qld.gov.au  Thu May  9 19:54:15 2002
From: Dennis_Martens at health.qld.gov.au (Dennis Martens)
Date: Fri, 10 May 2002 09:54:15 +1000
Subject: SUMMARY:  SGID on a directory, mv blows the theory!
Message-ID: <scdb98d1.035@HEALTH-ES2.HEALTH.QLD.GOV.AU>

Thanks to all who replied:

Spencer Hoffman
Dave Mitchell
Johan Hartenberg
Chris Hoogendyk
Kevin Reichhart
Michael Schulte
Matthew Stier
Darren Dunham
Jason Haynes
Jay Lessert
David Foster

All of whom pointed out the behaviour of mv opposed to cp. Thanks, but that's
what I had pointed to in the first place! More of a concern is that the stated
behaviour of a SGID (all files created in the directory will belong to the
named group) does not happen with a mv.

As you all stated, mv does not create a "new file", but modifies the pointer
in an existing inode. That's fine, but for all intents and purposes, it IS a
new file in THIS directory.

Johan had the best suggestion: write a wrapper script for mv. I will do this,
because the application that we are running is very reliant on file
ownerships.

Thanks for your time.


Original Post:

All

A small but very annoying problem. Nothing appropriate in the archives, so
would appreciate the help.

We have a directory on a Solaris8 server which has the SGID bit set, allowing
all files created in the directory to belong to the same group, the one that
the directory belongs to. This works fine when creating new files with touch,
cat, vi and so on, and works fine with cp as well. But when you mv a file,
the
rules seem to go out the window, and the file belongs to the default group of
the user doing the mv.

Is this a feature?

Thanks, I will send out a summary.







Dennis Martens
Unix Administrator, BCISU
Queensland Health , Brisbane Australia
617 32340729
Fax 32341289


**********************************************************************
This e-mail, including any attachments sent with it, is confidential
and for the sole use of the intended recipient(s). This confidentiality
is not waived or lost if you receive it and you are not the intended
recipient(s), or if it is transmitted/ received in error.

Any unauthorised use, alteration, disclosure, distribution or review
of this e-mail is prohibited.  It may be subject to a statutory duty of
confidentiality if it relates to health service matters.

If you are not the intended recipient(s), or if you have received this
e-mail in error, you are asked to immediately notify the sender by
telephone or by return e-mail.  You should also delete this e-mail
message and destroy any hard copies produced.
**********************************************************************
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


Dennis Martens
Unix Administrator, BCISU
Queensland Health , Brisbane Australia
617 32340729
Fax 32341289


**********************************************************************
This e-mail, including any attachments sent with it, is confidential
and for the sole use of the intended recipient(s). This confidentiality
is not waived or lost if you receive it and you are not the intended
recipient(s), or if it is transmitted/ received in error.

Any unauthorised use, alteration, disclosure, distribution or review
of this e-mail is prohibited.  It may be subject to a statutory duty of
confidentiality if it relates to health service matters.

If you are not the intended recipient(s), or if you have received this
e-mail in error, you are asked to immediately notify the sender by
telephone or by return e-mail.  You should also delete this e-mail
message and destroy any hard copies produced.
**********************************************************************
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From david at luyer.net  Thu May  9 21:02:13 2002
From: david at luyer.net (David Luyer)
Date: Fri, 10 May 2002 11:02:13 +1000
Subject: SUMMARY:  tmpfs bytes/inode ratio?
Message-ID: <000601c1f7be$52137ba0$638317d2@pacific.net.au>

The unfortunately correct answer (from Matthew Stier, and actually
straight from the tmpfs manpage on Solaris 8, where I'd missed it):

      Another constraint is that the number of files available  in
      a   tmpfs  file  system  is calculated based on the physical
      memory of  the  machine   and  not  the  size  of  the  swap
      device/partition.   If  you  have too many files, tmpfs will
      print a warning message and you will be unable to create new
      files. You cannot increase this limit by adding swap space.

(ie. even though I only need a small amount of RAM for the CVS to run,
and a small amount of temp space, I need to increase the physical RAM
to let me have enough lock files/directories on the tmpfs)

There was one suggestion of tunefs, however that can only tune "real"
partitions (ones that have some form of device at least) and only
tunes contiguous blocks, rotational delay, block grouping, free
space precentage and space vs time optimization.

Some suggested tmpfs didn't use inodes as it was memory based.  This
isn't correct - there are still inodes, even if they might not be in
the same format as UFS inodes.  tmpfs supports hard links, persistance
of deleted files, all the common Unix file system semantics, so it
has a form of inodes.

As to working out the number of inodes on tmpfs, none of the standard,
xpg4 or ucb df's show it (in 'df -o i'), however GNU df (in 'df -i')
does accurately report the number of inodes used and free on tmpfs.
I don't know why the Solaris df's have this flaw.

David.
--
David Luyer                                     Phone:   +61 3 9674 7525
Network Development Manager    P A C I F I C    Fax:     +61 3 9699 8693
Pacific Internet (Australia)  I N T E R N E T   Mobile:  +61 4 1111 BYTE
http://www.pacific.net.au/                      NASDAQ:  PCNTF
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From peter at ashlyn.co.uk  Fri May 10 07:21:51 2002
From: peter at ashlyn.co.uk (Peter Stokes)
Date: Fri, 10 May 2002 12:21:51 +0100
Subject: SUMMARY: What packages for SunPCi?
In-Reply-To: <3CDBA7E8.91640355@sun.com>
Message-ID: <FFEPIJGBHBKCGCNMMAGDAEAFDMAA.peter@ashlyn.co.uk>

Hi All

Quick responses as usual!!

Below is the most complete I received (Krister Bergman). Thanks to all that
replied with much the same info about download location.

Peter

>
> Hi
>
> Can anyone help me with what packages I need and where they can be found
for
> Sol 8 for the original SunPCi card fitted to an Ultra 5?

Assuming that "orignal SunPCi card" means the SunPCI or SunPCI+ (AMD
K6-2 300 or 400Mhz).

Latest version of software is 1.3 and package is called SUNWpci.

Download page:
http://www.sun.com/desktop/products/sunpci/sunpci_download.html

Necessary patches for Solaris 8
108652 -- all systems
108605 -- FFB systems only
108604 -- AFB systems only
108606 -- m64 systems only


Good luck

Krister Bergman
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From ahoesch at smartsoft.de  Fri May 10 09:27:04 2002
From: ahoesch at smartsoft.de (Andreas Hoeschler)
Date: Fri, 10 May 2002 15:27:04 +0200
Subject: Summary: Configuring timezone
Message-ID: <9E3B3B22-6419-11D6-B1B2-003065CCA582@smartsoft.de>

Hi all,

thanks to all that responded:

"Haywood, Steven" <shaywood at hurricaneseye.com>
Douglas Palmer <palmer at nyed.uscourts.gov>
Davorin Bengez <dbengez at interactive1.hr>
Casper Dik <Casper.Dik at Sun.COM>
Hendrik Visage <hvisage at is.co.za>
  "Glass, David (UDB)" <GlassD at bp.com>
krister bergman <krister.bergman at sun.com>

BTW: /etc/default/init would only get read again after a reboot.
thus to test the TZ variables, do something like:
# TZ=European/Timezone date
      ^^^^^^^^^^^^^^^^^ Replaced with the correct timezone (You might 
need to zic European or something equivalent)

The correct timezone for Germany appears to be CET. One further 
important thing to note is that /etc/default/init gets read during 
system boot only.

Thanks a lot,

    Andreas
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From Murtuza_Bekhushi at infy.com  Fri May 10 12:13:40 2002
From: Murtuza_Bekhushi at infy.com (Murtuza Bekhushi)
Date: Fri, 10 May 2002 21:43:40 +0530
Subject: SUMMARY: Domain name.
Message-ID: <F0CB3C9983B77E4AB4ADEFA63DAB109F2EE55C@twrmsg03.ad.infosys.com>

Hi all,

	The right solution was provided by many people.  Actually it is
an harmless error.

	The simplest solution was to put fully qualified domain name in
/etc/hosts file like

101.101.10.10   sun450e02   sun450e01.ad.infy.com
( First one should definitely be IP address,  second and third fields
can be any one of short name and FQDN)

	Previously in my /etc/hosts it had only IP address and short
Name, hence the error message.

	Some people also suggested to check out DNS server
connection/configuration. ( /etc/resolv.conf  and /etc/nsswitch.conf )
	Another solution is to put entry of domain name in
/etc/defaultdomain.

Thanks to following people for showing me the way to solutions.

Xu Ying, Krister Bergman, Jennifer Stults, Patrick Novak, Pierre
Zimmermann, Bill Fenwick, William Enestvedt, John Leadeham, Mike, David
Harrington, Todd Fiedler, Steve Mickeler, Phil Rainey, Dennis Peterson,
Randy Romero, Guido Kirschner, sysadmin at astro , Dana, Jason Haynes, Atul
Gore, ltiu, Tony Walsh, Martin Marshall, Sandip Sale, Nils Schoyen

Thanks and Regards,
Murtuza.

-----Original Message-----
From: Murtuza Bekhushi
Sent: Wednesday, May 08, 2002 7:09 PM
To: sunmanagers at sunmanagers.org
Subject: Domain name.

Hi friends,

	One of our Sun servers,   keeps on giving this message on
Console :

May  8 18:21:00 sun450e02 sendmail[6792]: unable to qualify my own
domain name (sun450e02) -- using short name

	Can anybody tell me whether it has any serious implications ?
Any way to resolve this ?
	Anyway, it has not given any problems in any way till date.

	It is SunOS 5.6 Generic_105181-23 sun4u sparc SUNW,Ultra-4

Regards,
Murtuza.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From cic_line at hotmail.com  Fri May 10 12:22:07 2002
From: cic_line at hotmail.com (CIC Line)
Date: Fri, 10 May 2002 12:22:07 -0400
Subject: SUMMARY: Commands to display all working variable values in
  /etc/system?
Message-ID: <F149NjGivzW8bts3QAs00014f36@hotmail.com>

Hi,
Thank you all for the super quick replies.

Thanks Andy Cordova's link that makes it clear:
http://www.princeton.edu/~unix/Solaris/troubleshoot/kerntune.html

Many people point to 'sysdef' which I actually tried before my posting but 
did not get what I want. (#sysdef | grep -i maxusers)

Fabrice Guerini & john65 wrote:
If you know your variable name in advance, you can do this:
# echo "maxusers/D" | adb -k /dev/ksyms /dev/mem

Beck, Joe wrote:
so, i guess the answer is no, cause to get the nic related stuff you have to 
use ndd & to other stuff you have to use modinfo or other variant specific 
cmds.

Original question:
Under Solaris, is there a command which can display all the current working 
values of variables in /etc/system such as maxusers=100, etc.?

Melissa Young
Unix System Administrator

_________________________________________________________________
Send and receive Hotmail on your mobile device: http://mobile.msn.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From return at trafficmagnet.net  Sun May 12 04:20:34 2002
From: return at trafficmagnet.net (Christine Hall)
Date: Sun, 12 May 2002 16:20:34 +0800
Subject: WWW.SUNMANAGERS.ORG
Message-ID: <200205122112.g4CLCsB12366@ns5.trafficmagnet.net>

An HTML attachment was scrubbed...
URL: http://www.sunmanagers.org/pipermail/summaries/attachments/20020512/ad6a567f/attachment.html 

From ador_sunga at iesingapore.gov.sg  Sun May 12 23:19:10 2002
From: ador_sunga at iesingapore.gov.sg (ador_sunga at iesingapore.gov.sg)
Date: Mon, 13 May 2002 11:19:10 +0800
Subject: SUMMARY:  SunFire V880
Message-ID: <OFF413D599.E503B409-ON48256BB8.000E5CA1@gems2.gov.sg>

Hi  Folks, Thanks for the following pple who takes time giving their
feedback :

Nelson.T.Caparroso
JC
Kevin Van Der Hart
Kanchan
Jed
Jon Jackson
Tim Chipman
Hichael Morton

The consensus is that V880 is the recommended model in terms of
scalability, reliability, durability and price cost.  The problem I've
mentioned about the "power supply" issue is already history, its already
been fixed for quite sometime (and to correct myself, the prob is actually
on  DC/DC  converter  and not the power supply).

Once again, Thanks for your contribution,

Ador Sunga
Information System Officer
IESingapore


Original Question:

Hello Fellow Admin,

I'd just like to know your experience on  this Sunfire V880 model in terms
of scalability .  We are about to purchase this unit but one of my
colleaque commented that modular processor/memory is not that good based on
his experience with his previous company. So I was tasked to make a
research before I can go ahead.

So far, I understant  that Sun makes a recall for their V880 because of the
power supply issue.  Aside from this, any good/bad experience that you can
share with me.

Thanks for your time, I'll summarise.
***************************************************************************

Ador,

   We have, at last count 6 V880's in house, and at least 2 more on order,
   and I
   must say that we've had nothing but excellent results. the modular CPU
   boards
   makes maintenance easier and faster in the event of a CPU replacement or
   install.
   The "recall"
   you mentioned was actually a mandatory field fix for a DC to DC
   converter,
   which is now really old news.  Our original two V800's where affected by
   this.
    It took the technician less than 10 minutes for each system to replace
   the
   affected component. All new V880's do not have this problem.

   For the price the V880's are a great bargain. From my experience they're
   fast,
   reliable and scalable.  You can replace I/O boards with the system
   running.
   They come with several lights out options and internal gigabit Ethernet.
   We
   obviously like them and will continue to purchase them.  In fact the
   next 2
   (and possibly four) will be clustered with Sun Cluster 3.0. (We also
   have a
   large range of other Sun systems from Netras to E10K's and 6800's.)

   JC

   ******************************************************************************

   Hello,

   We installed 2 V880's in February to run our ERP system and house all of
   our
   Oracle databases. They replaced an E4500 and an E5500 which cost 3 times
   as
   much as the V880. The V880 outperforms the older servers in every
   aspect.
   Sun did have the power supply recall and they have replaced ours.
   Actually
   it wasn't a power supply but a DC/DC converter that was installed on the
   CPU/Memory cards. The odds of having one of the early converters fail
   was
   only 1 in 1000 though. The only actual problem we have had was a DIMM
   that
   would give us correctable errors. Fortunately they were correctable and
   the
   server never crashed or rebooted. Other than that they have been working
   flawlessly for 3 months now. Hope this helps.

   ------
   Kevin Van Der Hart
   -----

   *******************************************************************************

   I recently evaluated one of the V880 and end up buying
   one, due to its scalability. Power supply issue has
   been fixed long back.

   Kanchan

   *******************************************************************************
   DC/DC convertor, you will not get one with a bad one since all inventory
   was replaced. The V880 is rock solid!

   Right now it supports 32G total memory, very soon 64G, after that 128G.
   The 900MHz Cheetah+ Cu processors are due in June. Great I/O, 9 PCI
   slots,
   redundant power (3 units/cords). 8 Processors now, with USIV you will be
   able to have 16 processors. This is a beast, a great machine. Buy two :)

   -jed

   ****************************************************************************

   I have a 6 way here that I'm pretty happy with... it's pretty much
   unloaded though, other then the occasional simulation from Cadence...

   ***************************************************************************

   We have 2 and love them....great performers.
   <jon.jackson>

   ***************************************************************************

   From what I can tell, the 880 is the most inexpensive 8-way-capable
   system
   that sun has produced.  I think previously, the "entry level" for 8-CPU
   system
   from sun was the e3500 (maxes out with 8 x 464mhz CPUs) and based on the
   price
   for the "bare box" and CPU,memory,expansion boards to upgrade to an
   8-cpu
   format, the curve comes out MUCH in favour for 880 over 3500 system.
   This is
   especially true since the 880 is using faster Ultra-3 CPUs yet the price
   point
   is better.

   In the long term, it is quite likely the 880 will accommodate 8 x
   1000mhz
   CPUs, and probably even a bit beyond that. (current standard format
   these days
   is with 750mhz CPUs I think).

   some rough comparison from purely a price perspective (*List prices*
   pulled
   from store.sun.com, info I previously had prepared ...)

   e3500 Server , 4 x 464mhz CPUs, 2 gigs Ram -> $104,850 USD

   V880 Server , 4 x 750 mhz Ultra-III Sparc CPUs, 8 gigs RAM, -> $49,995
   USD

   Upgrade costs:

   $15,000 USD list price  per 464mhz Ultra2 CPU for the e3500
           (plus extra for system boards and memory...)

   $23,000 USD list price for Dual-750mhz Ultra3 with 4 gig ram "Module"
   for v880

   (ie, includes system board, 2 CPUs, ram)


   Anyhow. Just some numbers to drive home the fact that the price-curve
   for the
   880 is more "in keeping" with 450 type "workgroup" servers rather than
   the
   "enterprise" (aka more expensive :-) hardware.


   If you are really concerned about scalability of a platform, it will
   involve
   some consideration of software, usage, "scaling model" overall, and
   certainly
   not just the "scalability" of a single given piece of hardware.


   Hope this is of slight use / interest,

   Tim Chipman

   **************************************************************************

   ador,

   The V880 must be ordered with 2 or 4 CPUs.
   You cannot add just CPUs; you must add the CPU & board.  This would
   apply to changing to higher speed CPUs also.


   Hope this helps,

   HM
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From brounb at adi-limited.com  Sun May 12 23:20:41 2002
From: brounb at adi-limited.com (Broun, Bevan)
Date: Mon, 13 May 2002 13:20:41 +1000
Subject: Summary: Veritas VM doesnt see Raid-5 set on A1000.
In-Reply-To: <20020513093248.B14860@adi-limited.com>; from
  brounb@adi-limited.com on Mon, May 13, 2002 at 09:32:48AM +1000
References: <20020513093248.B14860@adi-limited.com>
Message-ID: <20020513132041.H14860@adi-limited.com>

Thanks to Scott Howard <scott at doc.net.au> who said to run

vxdctl enable

The real problem is that Im a veritas newbie and should have RTFM. My
installation of veritas was not complete.

BB

on Mon, May 13, 2002 at 09:32:48AM +1000, Broun, Bevan <brounb at adi-limited.com> wrote:
> hi all
> 
> I have created a raid-5 set on an A1000. The format command shows the
> raid-5 set as 
> 
> c3t5d0 <Symbios-StorEDGEA1000-0301 cyl 65533 alt 2 hd 64 sec 169>
> 
> the vxdiskadm command shows me the underlying disk instead
> 
>   c3 c4t2:      all disks on both controller 3 and controller 4, target 2
>   c3t4d2:       a single disk
> 
> How do I get veritas to see c3t5d0 ?
> 
> Thanks in advance
> 
> BB
> _______________________________________________
> sunmanagers mailing list
> sunmanagers at sunmanagers.org
> http://www.sunmanagers.org/mailman/listinfo/sunmanagers
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From Martin.Gabel at ffm3.siemens.de  Mon May 13 05:08:59 2002
From: Martin.Gabel at ffm3.siemens.de (Gabel Martin)
Date: Mon, 13 May 2002 11:08:59 +0200
Subject: SUMMARY: redirecting /dev/console
Message-ID: <71BA7EC53702D3118E4500104BAFCDBE344C13@ffmr222a.ffm3.siemens.de>

On Wednesday 08 May 2002 00:21:
> Hi everybody !
>
> Our customer is running a e4000 and a few workstations (Ultra 1, 10 and
60)
> all with Solaris 2.5.1.
> The Server (e4000) is in the basement while the rest of the workstations
> are on the 3rd floor.
> Our customer is running some "experimental" client/server software which
> tent to crash...
> Anyway, my task is to check the applications on the workstation and, if
the
> server hangs, to reboot it.
> I can log in the server and do an init 6 - no problem. But I can't see how
> far the server is with shutting down and/or
> (more critical) with comming up. Several processes and database
> applications must be started.
> The point is: all system calls are directet to /dev/console (hope so...)
> and to the serial port. How can I
> redirect the output to one of the workstations ?? The server is not
running
> any kind of X-windows.
---------------------------------------------------------------
Thanks for all who have replied.

The problem is that I've not terminal server or cable which is long enough
to fit. (basement to 3rd floor  is aprox. 300m)
So I still have to go down in the basement and watch until the server comes
up ....

Anyway, Thank you verry much:

ltiu [ltiu at alumni.sfu.ca]
Jason Harkins [jharkins at sitesmith.com]
Steve Camp [steve at aslan.camp.com]
Matthew SAMS [maclean at cs.mcgill.ca]
Fricot, Daniel [daniel.fricot at barco.com]
j.huettemeister at t-online.de
Pierre Zimmermann [Pierre.Zimmermann at tecnomen.fi]
Callum Hughes [Callum.Hughes at sis.securicor.co.uk]
bergman at merctech.com

cu
may


martin.gabel at ffm3.siemens.de
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From mikelist at sky.net  Mon May 13 11:05:32 2002
From: mikelist at sky.net (Mike's List)
Date: Mon, 13 May 2002 10:05:32 -0500 (CDT)
Subject: SUMMARY: SDS RAID-5
Message-ID: <Pine.GSO.3.96.1020513100506.23497B-100000@pulsar.sky.net>

Well, I started the metainit command around 2pm and guess the RAID finished
sometime late at night or early morning.  By the time I left the office, it
wasn't finish, I came in the next morning, it's done.  So, doing a software
raid build on a 25G slice/3 disks takes quite a bit of time.



- Mike


-----Original Post-----

> How long does it take to make a RAID-5?  25G per slice, and my HD light
> just lit up for over two hours, I don't have my prompt back, the system
> seems froze? (Solaris 8 2/02 w/SDS that comes with the OS)
> 
> # metainit /dev/md/dsk/d13 -r /dev/dsk/c0t1d0s5 /dev/dsk/c1t1d0s5
>  /dev/dsk/c2t1d0s5 -i 8k
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From Ramiro.Santos at commerzbankib.com  Mon May 13 12:05:40 2002
From: Ramiro.Santos at commerzbankib.com (Santos, Ramiro)
Date: Mon, 13 May 2002 18:05:40 +0200
Subject: SUMMARY: How to interpret the prtconf output
Message-ID: <1D1B039860ABD41188C500508BAF2EB20248329B@xmx9fraib.fra.ib.commerzbank.com>

Hi, the solution is quite simple,
just try to read the isp man pages.

# man isp


Thanks to:
	Dave Mitchell
	John Riddoch



-----Original Message-----
From: Santos, Ramiro [mailto:Ramiro.Santos at commerzbankib.com]
Sent: Monday, May 13, 2002 3:01 PM
To: sunmanagers at sunmanagers.org
Subject: How to interpret the prtconf output


Hi people,
I am looking for some depth docs about the prtconf tool.
The problem is, I would like to find out in which mode the scsi drives are
running.

i. e.

# prtconf -v
	...
        scsi, instance #0 
            Driver properties: 
                name <target6-TQ> length <4> 
                    value <0x00000000>. 
                name <target6-wide> length <4> 
                    value <0x00000000>. 
                name <target6-sync-speed> length <4> 
                    value <0x00004e20>. 
                name <target5-TQ> length <4> 
                    value <0x00000000>.
	...

The output tells me the devices are "wide" but I have no idea
what "target6-sync-speed" means or the hex addresses.
What means "wide"? 40MB or 80MB?


Thank you in advance


Yours sincerely,

Ramiro Santos

ZIT P 7.41 Unix Server Implementation

_______________________________________________________

Contact:   Phone:     +49-(0)-69-136-43523
           Fax:       +49-(0)-69-136-47040
           E-Mail:    Ramiro.Santos at CommerzbankIB.com

Address:   Commerzbank AG
           Zentraler Servicebereich IT Production
           C/S Services Investment Banking - ZIT P 7.41
           Mainzer Landstr. 151 - DLZ 1, 06.52.0560
           D-60321 Frankfurt / Main
_______________________________________________________
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From carolynlee1999 at hotmail.com  Mon May 13 12:14:18 2002
From: carolynlee1999 at hotmail.com (Carolyn Lee)
Date: Mon, 13 May 2002 16:14:18 +0000
Subject: SUMMARY: a question of "find" command
Message-ID: <F93dzgXOoE4zyYJOGES000147ca@hotmail.com>

Hi Gurus,

Sorry for this a little late summary. Thank you all who responses for your 
help! I got more than 60 responses, I apologize I am not able to list all 
names who replied.

Quite a few people pointed out that I missed a "\{};" in the script. That is 
not the right answer to my question. Actually I knew that but I missed it 
when I send the request out.

i.e., the following commands which quite a few people suggested to try did 
not work if there are multi entries. The shell is expanding the $1 as it 
gets passed:

find . -name $1 -exec grep $2 {} \; -print
find . -name "$1" -exec grep "$2" {} \; -print
find . -name $1 -print | xargs grep $2


Larye D. Parkins pointed to the right direction. The following is from his 
message, that solved my problem:
--------------------------
find . -name \'$1\' -exec grep $2 {} \;

The escaped single quotes are needed to permit variable expansion by the 
shell for $1, but pass the "*" form (since in your example, $1 contains an 
escaped wild card) to 'find' without reexpansion: i.e, the shell will 
execute the find command as if you typed

  find . -name '*.h' ... at a shell prompt.

The results of the find are passed one at a time to the -exec target via the 
"{}", and you need the escaped semicolon to end the exec string.  If you 
need to know the file name in which the string was found, you need to use

        find . -name \'$1\' -exec grep -l $2 {} \;

which will output (example):
      signal.h
indicating the grep string was found in this file.

If you use
        find . -name \'$1\' -print -exec grep $2 {} \;
it will output (example):
        foo.h
        bar.h
        signal.h
        #define SIG_IGN   ...
        baz.h
which shows the names of all the files that did NOT contain the string, plus 
the strings found, after the file name(s) in which they were found
-----------------------------------------

Karl Vogel and Lee Trujillo sent me the user-friendly shell script, I tested 
it and it worked. I can send you that if you are interested in that.

Thanks again to all who replied,
- Carolyn -

>Hi Gurus,
>
>It might be a simple question, but I do not know how to do the trick.
>
>I want to write a shell script to use 'find' command to find some files and
>do a 'grep' on it, it works like this way:
># myfind "*.h" SIG_IGN
>
>The following is my script:
>#! /bin/sh
>/usr/bin/find . -name $1 -exec grep $2 -print
>
>It does not work, how can I do this?
>
>TIA for your help.
>


_________________________________________________________________
MSN Photos is the easiest way to share and print your photos: 
http://photos.msn.com/support/worldwide.aspx
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From nemo at circinus.com  Mon May 13 12:59:46 2002
From: nemo at circinus.com (Anton)
Date: Mon, 13 May 2002 09:59:46 -0700 (PDT)
Subject: SUMMARY: Netra X1 DMFE errors!
In-Reply-To: <Pine.GSO.4.43.0205122240260.16468-100000@zeta.circinus.com>
Message-ID: <Pine.GSO.4.43.0205130959180.19301-100000@zeta.circinus.com>

There is a patch for this at Sun's site... d'oh.

http://sunsolve.sun.com/pub-cgi/retrieve.pl?doc=fsrdb%2F43724&zone_128=DMFE&wholewords=on

On Sun, 12 May 2002, Anton wrote:

> Guys, I have a Netra X1 which give frequent errors in /var/adm/messages:
> Has anyone else seen this and how do I fix it?  Patch?
>
>
> May 12 03:40:26 zeta dmfe: [ID 693617 kern.info] NOTICE: DMFE: Abnormal
> interrupt, status 0xf
> c548800: Timer expired
> May 12 03:40:26 zeta dmfe: [ID 634597 kern.info] NOTICE: DMFE: CR0
> fe02e000
> May 12 03:40:26 zeta dmfe: [ID 634597 kern.info] NOTICE: DMFE: CR1
> ffffffff
> May 12 03:40:26 zeta dmfe: [ID 634597 kern.info] NOTICE: DMFE: CR2
> ffffffff
> May 12 03:40:26 zeta dmfe: [ID 634597 kern.info] NOTICE: DMFE: CR3
> 62019490
> May 12 03:40:26 zeta dmfe: [ID 634597 kern.info] NOTICE: DMFE: CR4
> 62005180
> May 12 03:40:26 zeta dmfe: [ID 634597 kern.info] NOTICE: DMFE: CR5
> fc548800
> May 12 03:40:26 zeta dmfe: [ID 634597 kern.info] NOTICE: DMFE: CR6
> 0224a007
> May 12 03:40:26 zeta dmfe: [ID 634597 kern.info] NOTICE: DMFE: CR7
> ffffa94b
> May 12 03:40:26 zeta dmfe: [ID 634597 kern.info] NOTICE: DMFE: CR8
> 00000000
> May 12 03:40:26 zeta dmfe: [ID 634597 kern.info] NOTICE: DMFE: CR9
> 010d83f0
> May 12 03:40:26 zeta dmfe: [ID 634597 kern.info] NOTICE: DMFE: CR10
> fffc0000
> May 12 03:40:26 zeta dmfe: [ID 634597 kern.info] NOTICE: DMFE: CR11
> fffe0000
> May 12 03:40:26 zeta dmfe: [ID 634597 kern.info] NOTICE: DMFE: CR12
> ffffff49
> May 12 03:40:26 zeta dmfe: [ID 634597 kern.info] NOTICE: DMFE: CR13
> 00000000
> May 12 03:40:26 zeta dmfe: [ID 634597 kern.info] NOTICE: DMFE: CR14
> 00000000
> May 12 03:40:26 zeta dmfe: [ID 634597 kern.info] NOTICE: DMFE: CR15
> 00000000
>
>

-- 
nemo at circinus.com
Nemo on EFNet IRC
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From sajeev20 at rediffmail.com  Mon May 13 13:29:54 2002
From: sajeev20 at rediffmail.com (sajeev  nv)
Date: 13 May 2002 17:29:54 -0000
Subject: SUMMARY:solaris 2.6 -to 8
Message-ID: <20020513172954.29149.qmail@webmail28.rediffmail.com>

An embedded and charset-unspecified text was scrubbed...
Name: not available
Url: http://www.sunmanagers.org/pipermail/summaries/attachments/20020513/82b27efa/attachment.ksh 

From payerle at benfranklin.physics.umd.edu  Mon May 13 13:50:49 2002
From: payerle at benfranklin.physics.umd.edu (Thomas M. Payerle)
Date: Mon, 13 May 2002 13:50:49 -0400 (EDT)
Subject: Summary: Anything special about partition #2
Message-ID: <200205131750.NAA32534@bofh.physics.umd.edu>

First off, thanks to the many prompt responses, including those who reminded
me that in Sunspeak it is "slice" #2 not partition #2.

The question was basically if one could redefine slice #2 of a disk to be
something other than the whole disk.

Though I didn't make an official tally, the responses overwhelming recommend
not touching it, though most responses in this vein sounded of religious
proscription ("thou shalt not mess with slice #2") rather than reasoned 
explanation.  One or two mentioned that some unspecified utilities may break
if it is changed, while another claimed Veritas and other disk management
software may break as use this partition to read the disk size.  One person
also claimed that it is used to access the initial cylinders for things like
partitioning a disk.

On the other hand, there were a few responses from adventurous spirits who
claimed that it is doable (at least for Solaris >= 2.6) and at least one person
who has successfully used it.  Most seem to believe that is a remnant from
pre-Solaris 2.6 days, and that basically the only thing preventing its use
is tradition and/or convention (coupled with the caveat that some utilities
may assume the tradition holds and get confused).

In short, a sharply divided response.  I'm inclined to side with the minority
(can be done, may break some questionably coded software).  However, as this
is a system disk, and may install Veritas at some point, and I don't absolutely
need all 8 partitions, I will probably take the advice of the majority and
leave it alone (for now, may play with it on data disks if need the extra
partition).

Thanks to all who responded, including:
Ray McCaffity, Callum Hughes, Vinnie German, Mike Salehi, Jonathon Hays,
Peter Kunst, Michael Kalus, Mark Neill, Deborah Santomauro,  Darren Dunham ,
Andrew_Rotramel,  Fabrice Guerini,  Randy Romero , David  Harrington,
Ramiro Santos,  Eric Priebe, and Jay Lessert (my apologies if missed anyone).


Tom Payerle 	
Dept of Physics				payerle at physics.umd.edu
University of Maryland			(301) 405-6973
College Park, MD 20742-4111		Fax: (301) 314-9525
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From jlkennedy at amcc.com  Mon May 13 17:04:23 2002
From: jlkennedy at amcc.com (Jeff Kennedy)
Date: Mon, 13 May 2002 14:04:23 -0700
Subject: SUMMARY: network stats
References: <3CDFFAE5.A70909F5@amcc.com>
Message-ID: <3CE02A57.4AE1196F@amcc.com>

I knew about netstat but it has no reporting capabilities like iostat;
'iostat -xc 2' to see what I mean.  I was looking for thruput statistics
of the kb per time increment variety, not merely packets.

It appears there is no native tool to do this, so ntop it must be.

Thanks to everyone who replied.  All but 2 gave some netstat option
(which did not give me what I was looking for), the other 2 said ntop.

~JK

Jeff Kennedy wrote:
> 
> I'm sure I'm just having a mis-fire of the synapses but I can't remember
> if it exists and I can't find it if it does.
> 
> Is there a native Solaris tool (pre-8) that will monitor/report network
> statistics similar to iostat?  I'm looking for something that will tell
> me how much data is passing across my gigabit interface per second.
> 
> Thanks.

-- 
=====================
Jeff Kennedy
Unix Administrator
AMCC
jlkennedy at amcc.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From Padraig.Lennon at Pioneerinvest.ie  Tue May 14 08:27:06 2002
From: Padraig.Lennon at Pioneerinvest.ie (Lennon, Padraig)
Date: Tue, 14 May 2002 13:27:06 +0100
Subject: Summary: Memory problems on E450
Message-ID: <99DABA002010D611A4DA00B0D068A5791E63FA@DUBSWAPM1>

Hi Gurus,
Sorry about the very late summary to this post, but i only got the problem
sorted out this weekend.

It turns out that the 3 DIMMS that had got to replace the failed ones had
done the rounds. One DIMM i have since discovered had been around the loop 3
times, been returned for repair, before being sent out to another customer.
What i did was to get SUN to pre-test two existing DIMMs for one week, after
which time i got them both sent to me. The second DIMM is backup for
possible transit damage.
I only got the proven DIMM installed last week and it is fine now. 
Keep the second DIMM on-site for a week afterwards for safety.

p.s. Be sure to stress test the memory for about 24hrs after changing using
SUN VTS. 
Also ensure that the machine is fully patched up.

Thanks to everyone that replied,

Hope this helps
-Padraig 
-----------------Original Post---------------------------

Hi Guru's,
I have a strange problem with one of our E450's.  About a month ago i
started getting the following errors in /var/adm/messages stating that
Memory module 1904 was experiencing memory problems

foo unix: [ID 908439 kern.notice] [AFT0] Multiple Softerrors:
foo unix: [ID 356634 kern.notice] 3 Intermittent, 253 Persistent, and 0
Sticky Softerrors accumulated
foo unix: [ID 340762 kern.notice] from Memory Module 1904

That seemed a straightforward error and i requested that a Sun Engineer come
and change the module. When he arrived he moved a known good module into
slot 1904, and placed the new module in 1901 (This was done to ensure that
it wasn't the slot that was causing the problem). This seemed fine and we
booted the machine up again and ran SunVTS stress test. 
The same errors occured again, but this time the errors were coming from
1901. We naturally thought that the dimm was bad and replaced this again,
this time placing 1804 into 1901 and the new DIMM in 1804 ( This was done to
rule out a faulty bank that was holding the 190x Dimms. 

We booted up again and all seemed fine. SUNvts passed with no errors, and we
left it and that. A day later though,  the problems started again - this
time from 1804. However the error messages were somwhat different

foo pcipsy: [ID 758641 kern.info]     AFSR=40830000.a4800000
AFAR=00000000.d0610fa8,
foo   double word offset=5, Memory Module 1804 id 4.
foo pcipsy: [ID 553544 kern.notice] syndrome bits 83
foo pcipsy: [ID 865758 kern.warning] WARNING: correctable error from pci0
(upa mid 4) during
foo DVMA read transaction

as well as:

foo unix: [ID 908439 kern.notice] [AFT0] Multiple Softerrors:
foo unix: [ID 356634 kern.notice] 3 Intermittent, 253 Persistent, and 0
Sticky Softerrors accumulated
foo unix: [ID 340762 kern.notice] from Memory Module 1804

I got onto SUN support who told me it looked like a motherboard error. We
changed the motherboard, and again SUNvts passed all tests. 

To my disgust the errors are back again. I have run SUN explorer on the host
a number of times which SUN have analysed, but have found no problems. Their
suggestion now is to break the memory interleave, disable a bank at a time
to try isolate the problem. I can't do this however as it is a production
host and all 4gb of memory is needed. 

I have search extensively in Sunsolve etc.. for clues but to no avail. I did
notice however that some people have had problem with E450's incorrectly
diagnosing a failed DIMM. prtdiag does not show any errors at all.

Has anyone come across a problem like this before, and if so what was the
cause?

E450 spec -- 4x480mhz processors, 4gb Mem ( interwoven)
Solaris 8 Patch 108528-12
Sunvts version 4.6

I will of course summerize no matter what the outcome.
Thanks
-Padraig
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From ndo at woup.net  Tue May 14 08:41:03 2002
From: ndo at woup.net (Nicolas Dorfsman)
Date: Tue, 14 May 2002 14:41:03 +0200
Subject: SUMMARY: Networker vs Netbackup
Message-ID: <01c301c1fb44$9d5d9b70$9001020a@toulouse.datamedia.fr>

    Sorry for the very late summary !  I hope to have many many
answers....not so much !

    Thanks to :
    Marco Greene
    Fabrice Guerini
    Mark Hargrave
    Stuart Whitby


        Most importants points to considere :
            - as usual scalability, performance, ease of admin and install,
etc...
            - ability to back up the most amount of data in a set timeframe
            - support for any platform
        Depending on requirements:
            - format of storage (tar, proprietary, etc)
            - ease and adaptability of scheduling
            - ease of scripting


    NetWorker vs Netbackup :
        - Some found NetBackup interface more intuitive. Some others not.
Hard to have a consensus.
        - NetWorker seems to have new functionalities sooner than NetBackup
        - NetBackup could store data on tapes with a non-proprietary format.
        - NetWorker is easier to script

    Last (but not the least), if you need to have backup/restore interface
built for users, NetWorker is the way to go. On the other hand if you need
someting much more for operators, NetBackup is a good choice.   I mean, with
NetWorker users could do their job without any operator assistance; but
NetBackup interface may be much more intuitive for operators.


        Thanks.

        Nicolas Dorfsman
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From b.hswe at genaissance.com  Tue May 14 12:57:19 2002
From: b.hswe at genaissance.com (Hswe, Barbara)
Date: Tue, 14 May 2002 12:57:19 -0400
Subject: SUMMARY: permissions on a directory
Message-ID: <3C5D9F2FC06FA849A9348AAE099617861AB8CB@SGENUS03.gpi.genaissance.com>

SETFACL was the solution and has been implemented successfully.  THANK YOU to
the following:

Alex Stade
Michael Schulte
Nicholas Dorfsman
Ron Dinwiddie
Dave Mitchell
Mitchell Baker
Adam Kirby
Sean Berry
Jay Koonz
Mark Bergman


-----Original Message-----
From: Hswe, Barbara
Sent: Tuesday, May 14, 2002 11:38 AM
To: Sunmanagers (E-mail)
Subject: permissions on a directory


Hello Managers,

I've searched high and low and tried all possible combinations, but I am
still
stumped.  I need to set permissions on a folder that temporarily give a
handful of users the ability to write to a folder that is owned by a certain
user.  I can't change the group permissions, because the group can't write,
and the other can't even be allowed to see the folder.   So, the only
alternative is to give the users the ability to read, write and delete as the
user who owns the folder.

Am I making this harder, or is there no solution?  setgid, setuid, and sudo
don't seem to apply, unless sudo can be assigned to a user other than root.

Barbara
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From pmora at cgob.junta-andalucia.es  Tue May 14 13:43:14 2002
From: pmora at cgob.junta-andalucia.es (Patricio Mora)
Date: Tue, 14 May 2002 19:43:14 +0200
Subject: SUMMARY: SSA 110 fans failed
Message-ID: <3CE14CB2.7AC56135@cgob.junta-andalucia.es>

No answer: no solution. Pray for your filesystems if a dualed SSA fan breaks. Thanks God, we only lost networker indexes. Many thanks anyway.

------- Original message --------

It's almost solutioned, but I've serious doubts about what was going on on my e10k domain. There are two
dual controler SSAs (every fs mirrored with VM on the contrary SSA), an A1000, and the boot RSM attached.
Yesterday from 22h to 00:45 these messages where continously appearing:

May  9 10:06:11 pepe unix: ID[SUNWssa.soc.link.1010] soc1: port 0 WWN 8a1553: message: Fans have failed!
Expect over temperature condition soon

At 00:55 oracle stopped itself saying it can't write something to disk. When we get in at 7 this morning,
filesystems in RSM and A1000 where responding normally, but even an ls on an SSA fs hung. We had to hard
stop (~#) because the domain didn't sync, and boot, fsck,...

What happened?, why the loss of the active controller or a power failure in an SSA is less harmful (VM
disabling mirrors) than the situation described? How can we prevent ourselves from this happening again?

Any hint will be much appreciated, many thanks to all of you.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From rmcewan at netopia.com  Tue May 14 14:16:58 2002
From: rmcewan at netopia.com (Ryan McEwan)
Date: Tue, 14 May 2002 12:16:58 -0600
Subject: Summary: Problem with jumpstart
In-Reply-To: <PGEOIKLFLPBCMDPOKBPBIEODMBAA.rmcewan@netopia.com>
Message-ID: <PGEOIKLFLPBCMDPOKBPBKEPKMBAA.rmcewan@netopia.com>

First off, many thanks to those that offered assistance.
Jeff Kennedy, Anwar Saadeh, Todd Urie, Gert-Jan Hanenaars, Robert Legate and
Steve Aizic.  I hope I did not miss anyone.

Anyways, after many many emails back and forth to these folks offering
suggestions such as snoops, double checking configs, etc, I began to notice
that the problem may actually be with the tftp server on the boot server.
Since I have tftp configured to run out of inetd, I hup'ed the inetd
process, but with no success.  I also as per one suggestion killed and
restarted the rarpd processes.  Once again to no avail.  I finally took the
windows approach and rebooted the boot server and that is what ultimately
fixed the problem.  Still not sure exactly what caused it, but as some have
said before, when in doubt, reboot.  It worked for me.
Thanks again.

Ryan
-----Original Message-----
From: Ryan McEwan [mailto:rmcewan at netopia.com]
Sent: Tuesday, May 14, 2002 10:14 AM
To: Sun Managers
Subject: Problem with jumpstart


Solaris 8 07/01.  I have a previous jumpstart env all configured and was
working great, until I am trying to jumpstart 3 more Netra T1 AC200 servers.
I have added entries in /etc/ethers and also /etc/bootparams as I have
always done in the past when it has worked.

Here is all I get.

ok boot net - install
Boot device: /pci at 1f,0/pci at 1,1/network at c,1  File and args: - install
Using Onboard Transceiver - Link Up.
Setting Half duplex mode.
Timeout waiting for ARP/RARP packet
Timeout waiting for ARP/RARP packet
Timeout waiting for ARP/RARP packet
Timeout waiting for ARP/RARP packet



How do I troubleshoot this to try and figure out why it is not picking up
the MAC address and beginning the install?  I have been able to successfully
jumpstart Netra T1 AC200's in the past, but just not these 3 new ones.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From Charles.Homan at GDC4S.Com  Tue May 14 15:22:39 2002
From: Charles.Homan at GDC4S.Com (Homan, Charles (NE))
Date: Tue, 14 May 2002 15:22:39 -0400
Subject: SUMMARY: How do you copy a socket?
Message-ID: <2575327B6755D211A0E100805F9FF9540BB291AB@ndhmex02.ndhm.gsc.gte.com>

Thanks for all of the responses.  There were four basic answers provided:

1. cpio - refused to copy the file
2. ufsdump - copied the file as a regular file (-rwxrwxrwx) instead of as a
socket (srwxrwxrws)
3. dd - didn't try

But the correct answer (provided first by Nicolas Dorfsman) was:
4. "The software which [created/listens to] this socket will re-create it."

I just booted the machine and sure enough NDS re-created its sockets, and
seems happy about it.

Thanks to:
Nicolas Dorfsman
Darren Dunham
Steve Elliott
Mark Hargrave
JULIAN, JOHN C
Sean Berry
Urie, Todd
Lennon, Padraig
TRUCKS, JESSE
Graham Wood
Jonathan Andrews
Bertrand_Hutin
Lars Hecking
Jeff Claunch
Steven Aizic
Fabrice Guerini

Regards,
Charles

>  -----Original Message-----
> From: 	Homan, Charles (NE)  
> Sent:	Tuesday, May 14, 2002 11:13 AM
> To:	'sunmanagers at sunmanagers.org'
> Subject:	How do you copy a socket?
> 
> OK, here's what I hope is an easy one: I am trying to move /var onto its
> own partition on the boot disk.  I have created a new partition in the
> blank space, newfs'ed it, booted from CD.  Then I mount and move stuff
> like so:
> 
> mount /dev/dsk/c0t0d0s0 /a
> mount /dev/dsk/c0t0d0s7 /a/newvar
> cd /a/var
> tar cf - . | ( cd /a/newvar ; tar xf - )
> 
> Should be no issue, but it seems there are two socket files (created by
> NDS) in /var, which tar barfs on.  One is:
> 
> srwxrwxrwx 1	root	other 0	May 9 9:13 s-n4ucmd
> 
> The other is similar.
> 
> How can I either copy over or re-create this socket file on the new
> partition?  Any help would be vastly appreciated!
> 
> Thanks!
> Charles
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From foster at dim.ucsd.edu  Tue May 14 15:39:10 2002
From: foster at dim.ucsd.edu (David Foster)
Date: Tue, 14 May 2002 12:39:10 -0700 (PDT)
Subject: [SUMMARY] Upgrade of sparcstation-10 from Solaris 2.5.1 -> 8
  (PROM  issues etc)
Message-ID: <200205141939.g4EJdAw28963@dim.ucsd.edu>

Question: Am I limited to 1GB/2GB root partition on a Sparc-10?

Consensus was that you're limited to a 2GB partition, as there is
no 3.x PROM version for the sun4m architecture. The FAQ states that
with PROM 2.5 or less you're limited to 1GB. 

I received a freebie upgrade to 2.25, and have configured / to be 1.2GB 
and have not had any problems, other than that it wants to boot from 'net' 
no matter what I set for "boot-device"; I have to explicitly "boot disk" (my
aliases for disk and disk0 are correct).

Thanks to:

Matthew Stier
Darren Dunham
Dragon King

> 
> I need to upgrade a Sparcstation-10 from Solaris 2.5.1 to 8.
> PROM version is 2.12, and according the FAQ (5.60) I see that with this
> older PROM I'm limited to a 1GB root partition.
> 
> My question is, if I upgrade the PROM will I still be limited
> to, say, 2GB for root partition? I would like to us a 4GB system
> disk. I've checked the archives.
> 
> Any other gotchas anyone can share with me for this upgrade?
> Currently using a GX frame-buffer, which is supported under
> Solaris 8 (I've been bitten by that before...oh gee we don't
> support that fb anymore).
> 
> Thanks, summary to follow.



   << All opinions expressed are mine, not the University's >>

  =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
   David Foster    National Center for Microscopy and Imaging Research
    Programmer/Analyst     University of California, San Diego
    dfoster at ucsd.edu       Department of Neuroscience, Mail 0608
    (858) 534-7968         http://ncmir.ucsd.edu/
  =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=

   "The reasonable man adapts himself to the world; the unreasonable one
   persists in trying to adapt the world to himself.  Therefore, all progress
   depends on the unreasonable."   -- George Bernard Shaw
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From rfransix at yahoo.com  Tue May 14 16:56:11 2002
From: rfransix at yahoo.com (pelicancomputers)
Date: Tue, 14 May 2002 13:56:11 -0700 (PDT)
Subject: SUMMARY: lo0 loopback usage explained
In-Reply-To: <3BD8AA3B9C18D34BA5099929909CFA05418448@m0319p35.nordstrom.net>
Message-ID: <20020514205611.79450.qmail@web14803.mail.yahoo.com>

thanks everyone for your time.

> Well ... any network traffic destined to the local
> host is going to go over the loopback, even if you
> try to use the ip on another interface ...
> 
> -----Original Message-----
> From: pelicancomputers [mailto:rfransix at yahoo.com]
> Sent: Monday, May 13, 2002 9:41 AM
> To: sunmanagers at sunmanagers.org
> Subject: lo0 loopback usage explained
> 
> 
> can you explain why the lo0 interface has such a
> high usage? where is all this usage going to? should
> it be corrected? how?
> 
> # netstat -rn
> 
> Routing Table: IPv4
>   Destination           Gateway           Flags  Ref
>  
> Use   Interface
> -------------------- -------------------- -----
> -----
> ------ ---------
> 66.111.176.16        66.111.176.18         U       
> 1 
>   598  eri0
> 224.0.0.0            66.111.176.18         U       
> 1 
>     0  eri0
> default              66.111.176.17         UG      
> 1 
>  6413  
> 127.0.0.1            127.0.0.1             UH     
> 397998691  lo0
LAUNCH - Your Yahoo! Music Experience
http://launch.yahoo.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From rfransix at yahoo.com  Tue May 14 17:36:34 2002
From: rfransix at yahoo.com (pelicancomputers)
Date: Tue, 14 May 2002 14:36:34 -0700 (PDT)
Subject: SUMMARY: kern warning explained
In-Reply-To: <20020514205246.68686.qmail@web14805.mail.yahoo.com>
Message-ID: <20020514213634.81475.qmail@web14806.mail.yahoo.com>

this could be due to some hardward changes recently
made on this server; the mouse or keyboard cable is
loose; another usb cable is loose or added hot; a
drive error. if you can't identify it, may i suggest
that you do the following:

reboot with the following command:

   mkdir -p /etc/backup
   mv /etc/path_to_inst* /etc/backup
   reboot -- -arv

Respond "y" to the question "Do you want to rebuild
path_to_inst?".

also, there is a patch that has this error fixed:
#109896.

--- pelicancomputers <rfransix at yahoo.com> wrote:
> Apr 17 12:06:17 myserver9393 unix: [ID 882636
> kern.warning] WARNING: interrupt level 9 not
> serviced
> 
> what's this all about? on a solaris 8 server.
> 
> thanks.
> LAUNCH - Your Yahoo! Music Experience
> http://launch.yahoo.com
> _______________________________________________
> sunmanagers mailing list
> sunmanagers at sunmanagers.org
>
http://www.sunmanagers.org/mailman/listinfo/sunmanagers
LAUNCH - Your Yahoo! Music Experience
http://launch.yahoo.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From mikelist at sky.net  Tue May 14 20:32:53 2002
From: mikelist at sky.net (Mike's List)
Date: Tue, 14 May 2002 19:32:53 -0500 (CDT)
Subject: SUMMARY: chown script
In-Reply-To: <Pine.GSO.3.96.1020514175343.19302C-100000@pulsar.sky.net>
Message-ID: <Pine.GSO.3.96.1020514193204.13108B-100000@pulsar.sky.net>

The below did the job (ie. where user1 owns /var/mail/user1 file)

for i in *; do chown $i $i; done

Thanks.


- Mike


On Tue, 14 May 2002, Mike's List wrote:

> Anyone has a basic script to change user's permission or know if it's in
> the archive...
> 
> ie. chown user1 user1 (w/out changing the . and ..)
>     chown user2 user2 ...and so on, I guess something like for $1=user
>     chown $1 $1 or something...
> 
> 
> Thanks.
> 
> 
> - Mike
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From murat.bayrakci at siemens.com.tr  Wed May 15 02:35:12 2002
From: murat.bayrakci at siemens.com.tr (Murat BAYRAKCI)
Date: Wed, 15 May 2002 09:35:12 +0300
Subject: SUMMARY:  L9 Autoloader management software
Message-ID: <5.1.0.14.2.20020515091103.00be3650@ankb001a>

My Original Question was :

At 11:08 14/05/02 +0300, you wrote: 
Hi managers,
Does anyone know any free command line software or gui to manage 
robot in L9 other than Networker or Solstice Backup ?


I think   http <http://mtx.sourceforge.net/>  ://mtx.sourceforge.net/
<http://mtx.sourceforge.net/>     has all the answers...   There is no
free GUI..



I  would like to thank to Mark ( Bergman? ),  Scott Howard ,Ric
Anderson, Jim S, Sal Serafino, Tim Hespe  and  Jay Lessert ..
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From scotrn at cox.net  Wed May 15 09:53:41 2002
From: scotrn at cox.net (scotrn)
Date: 15 May 2002 09:53:41 -0400
Subject: SUMMARY: Loading perl modules bassed on Solaris uname
In-Reply-To: <1021406764.1227.67.camel@sywn001.engrs.infi.net>
References: <1021406764.1227.67.camel@sywn001.engrs.infi.net>
Message-ID: <1021470821.1263.23.camel@sywn001.engrs.infi.net>

Thanks to those who responded. 

There are a few discussons on the net about this. Apparently the use 
statements are processed before ANY other perl code so... 

use lib $path is equal to use lib "NULL". 
You cannot use a variable yet because perl has not parsed that bit of
code that defines $path. 

The only way I am able to push a dynamic directory onto the INC array is
by wrapping perl in a shell script setting LD_LIBRARY_PATH or 
by adding a bit of ugly code. Still can't seem to get DynaLoader to look
in a different directory so I created a sun4-solaris/auto directory under 
each 5.6,5.7 amd 5.8 directory to hold my .so files  

use strict;
use POSIX qw(uname);
my ($uname_s, $uname_r)  = (POSIX::uname())[0,2];

if (($uname_s eq "SunOS") && ($uname_r eq 5.6)) {
      use lib qw(/var/opt/modules/SunOS/5.6);
   } elsif (($uname_s eq "SunOS") && ($uname_r eq 5.7)) {
      use lib qw(/var/opt/modules/SunOS/5.7);
   } elsif (($uname_s eq "SunOS") && ($uname_r eq 5.8)) {
      use lib qw(/var/opt/modules/SunOS/5.8);
   } else {
      die("$0 is only supported on Solaris 2.6, 2.7 & 2.8 \n");
   }

Scot 

On Tue, 2002-05-14 at 16:06, scotrn wrote:
> Hi; 
> 
>  I was wondering if any of you have tried to dynamicly load 
> a module from a directory based on the rev of OS? 
> 
> In other words....
> my ($uname_s, $uname_r) = split('\s',`uname -sr`);
> 
> use lib "/share/local/modules/$uname_s"."/"."$uname_r";
> 
> This seems to work but modules which use DynaLoader to load shared
> objects it's falling back to the original @INC and not looking in 
> my lib path. 
> 
> Trying to support an NFS share which has multiple versions of compuled 
> modules bassed on OS Rev. 
> 
> Thanks 
> Scot 
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From craigg at psu.edu  Wed May 15 10:02:12 2002
From: craigg at psu.edu (Craig L. Gruneberg)
Date: Wed, 15 May 2002 10:02:12 -0400
Subject: SUMMARY: UFSDUMP and single files
Message-ID: <3CE26A64.FBF24C9C@psu.edu>

Well, I thought Buddy Lumpkin was onto something when he suggested the
following command syntax:

ufsdump 0f /dev/rmt/0cbn /export/home ./foo1/foo2/test.ksh

when the path to the single file wanted is:

/export/home/foo1/foo2/test.ksh

but that dumps the entire filesystem at /export/home.

One person suggested that ufsdump was strictly a "ufs filesystem" dump
but that is not correct as I can dump single files that are located
elsewhere on my Sun, i.e.:

ufsdump 0f /tmp/dumpfile /opt/SUNWspci2/drivers/win98/sisaudio/_sys1.cab

works just fine.

I still cannot dump my SunPCi image which is at:

/zoso/pc_images_backup/C.diskimage

{clg} zoso: ufsdump 0f /tmp/dumpfile /zoso/pc_images_backup/C.diskimage
  DUMP: Writing 32 Kilobyte records
  DUMP: Date of this level 0 dump: Wed May 15 09:58:27 2002
  DUMP: Date of last level 0 dump: the epoch
  DUMP: Dumping /dev/rdsk/c1t1d0s5 (zoso:/zoso) to /tmp/dumpfile.
  DUMP: Mapping (Pass I) [regular files]
  DUMP: Cannot find filesystem mount point for
`/zoso/pc_images_backup/C.diskimage'
  DUMP: The ENTIRE dump is aborted.

where zoso is:

/dev/dsk/c1t1d0s5    29160814 26983272 1885934    94%    /zoso

Thanks to all who responded to my plea for help! Best I can do is:

ufsdump 0f /tmp/dumpfile /zoso/pc_images_backup/

which gets the complete directory where C.diskimage is stored.

-craig
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From scotrn at cox.net  Wed May 15 10:12:11 2002
From: scotrn at cox.net (scotrn)
Date: 15 May 2002 10:12:11 -0400
Subject: UN-SUMMARY: Loading perl modules bassed on Solaris uname
In-Reply-To: <1021470821.1263.23.camel@sywn001.engrs.infi.net>
References: <1021406764.1227.67.camel@sywn001.engrs.infi.net> 
  <1021470821.1263.23.camel@sywn001.engrs.infi.net>
Message-ID: <1021471931.1261.27.camel@sywn001.engrs.infi.net>

Well I thought it was working but since I have compiled and installed
the modules and libs into their respective directories perl is now
loading ALL the lib paths disregarding all if statements. 

@#^^$&$@7 

On Wed, 2002-05-15 at 09:53, scotrn wrote:
> Thanks to those who responded. 
> 
> There are a few discussons on the net about this. Apparently the use 
> statements are processed before ANY other perl code so... 
> 
> use lib $path is equal to use lib "NULL". 
> You cannot use a variable yet because perl has not parsed that bit of
> code that defines $path. 
> 
> The only way I am able to push a dynamic directory onto the INC array is
> by wrapping perl in a shell script setting LD_LIBRARY_PATH or 
> by adding a bit of ugly code. Still can't seem to get DynaLoader to look
> in a different directory so I created a sun4-solaris/auto directory under 
> each 5.6,5.7 amd 5.8 directory to hold my .so files  
> 
> use strict;
> use POSIX qw(uname);
> my ($uname_s, $uname_r)  = (POSIX::uname())[0,2];
> 
> if (($uname_s eq "SunOS") && ($uname_r eq 5.6)) {
>       use lib qw(/var/opt/modules/SunOS/5.6);
>    } elsif (($uname_s eq "SunOS") && ($uname_r eq 5.7)) {
>       use lib qw(/var/opt/modules/SunOS/5.7);
>    } elsif (($uname_s eq "SunOS") && ($uname_r eq 5.8)) {
>       use lib qw(/var/opt/modules/SunOS/5.8);
>    } else {
>       die("$0 is only supported on Solaris 2.6, 2.7 & 2.8 \n");
>    }
> 
> Scot 
> 
> On Tue, 2002-05-14 at 16:06, scotrn wrote:
> > Hi; 
> > 
> >  I was wondering if any of you have tried to dynamicly load 
> > a module from a directory based on the rev of OS? 
> > 
> > In other words....
> > my ($uname_s, $uname_r) = split('\s',`uname -sr`);
> > 
> > use lib "/share/local/modules/$uname_s"."/"."$uname_r";
> > 
> > This seems to work but modules which use DynaLoader to load shared
> > objects it's falling back to the original @INC and not looking in 
> > my lib path. 
> > 
> > Trying to support an NFS share which has multiple versions of compuled 
> > modules bassed on OS Rev. 
> > 
> > Thanks 
> > Scot 
> _______________________________________________
> sunmanagers mailing list
> sunmanagers at sunmanagers.org
> http://www.sunmanagers.org/mailman/listinfo/sunmanagers
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From david_luyer at pacific.net.au  Thu May  9 20:59:39 2002
From: david_luyer at pacific.net.au (David Luyer)
Date: Fri, 10 May 2002 10:59:39 +1000
Subject: SUMMARY:  tmpfs bytes/inode ratio?
Message-ID: <000501c1f7bd$f651cc40$638317d2@pacific.net.au>

The unfortunately correct answer (from Matthew Stier, and actually
straight from the tmpfs manpage on Solaris 8, where I'd missed it):

      Another constraint is that the number of files available  in
      a   tmpfs  file  system  is calculated based on the physical
      memory of  the  machine   and  not  the  size  of  the  swap
      device/partition.   If  you  have too many files, tmpfs will
      print a warning message and you will be unable to create new
      files. You cannot increase this limit by adding swap space.

(ie. even though I only need a small amount of RAM for the CVS to run,
and a small amount of temp space, I need to increase the physical RAM
to let me have enough lock files/directories on the tmpfs)

There was one suggestion of tunefs, however that can only tune "real"
partitions (ones that have some form of device at least) and only
tunes contiguous blocks, rotational delay, block grouping, free
space precentage and space vs time optimization.

Some suggested tmpfs didn't use inodes as it was memory based.  This
isn't correct - there are still inodes, even if they might not be in
the same format as UFS inodes.  tmpfs supports hard links, persistance
of deleted files, all the common Unix file system semantics, so it
has a form of inodes.

As to working out the number of inodes on tmpfs, none of the standard,
xpg4 or ucb df's show it (in 'df -o i'), however GNU df (in 'df -i')
does accurately report the number of inodes used and free on tmpfs.
I don't know why the Solaris df's have this flaw.

David.
--
David Luyer                                     Phone:   +61 3 9674 7525
Network Development Manager    P A C I F I C    Fax:     +61 3 9699 8693
Pacific Internet (Australia)  I N T E R N E T   Mobile:  +61 4 1111 BYTE
http://www.pacific.net.au/                      NASDAQ:  PCNTF
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From nschelle at crosskeys.com  Wed May 15 12:30:07 2002
From: nschelle at crosskeys.com (Neil Schellenberger)
Date: Wed, 15 May 2002 12:30:07 -0400
Subject: [SUMMARY] OBP setting for full duplex NIC?
References: <15575.65145.93257.995925@crosskeys.com>
Message-ID: <15586.36111.816175.60771@crosskeys.com>

Folks,

Briefly, the original question was:

  Is there any way to set the hme link mode (duplex) at the OBP level?

There were only three responses, one of which was a clipping of the
FAQ material on what to set in /etc/system, which isn't directly
germane to this problem.

Darren Dunham helpfully pointed out a twist that I hadn't considered:

    > I can't simply change the boot image /etc/system because a) that won't
    > help with the tftp portion of the boot, and b) I need to be able to
    > jumpstart both half and full-duplex machines.  I suppose I could
    > maintain two boot images, but that seems like overkill (as well as
    > being error prone).

    You will need to anyway.  While an OBP command would probably help the
    TFTP portion, once the kernel loads the hme driver, it will probably
    completely reset the interface, and redo negotiation (unless set).

    If at all possible, I recommend staying away from /etc/system mods and
    instead put in an explicit rc script that ndds the interfaces.  Such a
    script could have logic in it to do its work only on a set of machines
    that you need it to (perhaps identified by hostname or IP).

Matthew Stier kindly pointed me at:

    <URL:http://www.cisco.com/warp/public/473/12.html>

Unfortunately, this material doesn't really help in this particular
situation.  During further discussion with him, he opined:

    As to your problem, I'd forget using full-duplex.  Before upgrading our
    network to 4000's, we used to have 5000's with these cards, and just
    locked them at half-duplex.

    Although Cisco support full-duplex on it's 10Base-T ports, 98% of all
    NICs do not.  It is a hardware limitation. (To support full-duplex,
    CSMA/CD must be disabled; and old 10Base-T designs did not permit this.)

    Full Duplexing did not become popular, or even a part of the standard
    until the ratification of the 100Base-T standards, and it was hapazardly
    backfitted onto 10Base-T cards. (Actually, I believe that those cards are
    really 10/100 cards with the 100Base-T feature permanently disabled.)

The only other piece of infomation I've managed to find is how to set
the link speed at the OBP via nvedit:

    probe-all install-console banner
    apply transfer-speed=10 /path/to/hme/device

Unfortunately, there doesn't seem to be an equivalent setting for link
mode.  Sigh.

Basically, it seems that you're SOL before /etc/system or /etc/rc?.d
are read.  Even then, in order to have a single jumpstart miniroot,
you'd need to have some sort of external logic for an rc script to
decide whether a given hosts should be full or half so that it can ndd
appropriately.  All of this seems predjudicial to having a simple,
self-maintaining, automatic jumpstart system in this environment.  It
seems that the sysadmin will need to manually change the Cisco port
configuration while machines are jumpstarting and then switch it back
again after.


Regards,
Neil

-- 
Neil Schellenberger          | Voice : (613) 599-2300 ext. 8445
Orchestream Americas Corp.   | Fax   : (613) 599-2330
350 Terry Fox Drive          | E-Mail: nschellenberger at orchestream.com
Kanata ON, Canada, K2K 2W5   | URL   : http://www.orchestream.com/
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From joanna.sizemore at grote.com  Wed May 15 11:19:30 2002
From: joanna.sizemore at grote.com (Sizemore, Joanna)
Date: Wed, 15 May 2002 10:19:30 -0500
Subject: SUMMARY:  Backup Hardware
Message-ID: <9F4F1AE59A6BD311B7DE000629507AC9028DE530@MA6>

Thank to all for the advice and recommendations.  Especially:  Jim Southerd,
Jeff Kennedy, Moti, Jed Dobson, Marcelino Mata, Penney Jaye Deeney,
Christophe Dupre, Andrew Stueve,  David Foster (who included a very
informative white paper that I will share if anyone wants a copy), Vincent
McIntyre, Raghu, Stuart Whitby.  I have a lot of information to digest but
thought I would go ahead and post a summary.  No clear cut answers, but good
information none the less so I have opted to post the responses instead of
trying to summarize.
_____

Not familiar with various media, but one thing I know is that Veritas has
agent for Solaris as well.

So you can run the Veritas backup on an NT and it will back up all your
Microsoft and Unix flavor files.
The Unix agent actually allows you to specify which partitions and
directories on the Unix box you want backed up.
_____
I am using two qualstar libraries
one with 100 tape slots and 6 ait2 tapes and one with 12 slots and two ait2
tapes .
I like both the performance and ( so far 2+ yrs ) 0 maintenance .
I would go with ait as it has proven to be fast and reliable.
as for the old dlt's just keep a system on the side for restores ....
just my 0.2 cents
_____
How many servers, how much storage? I would go for a mixed-media library
and use DLT & a new technology. AIT and LTO are the big ones. I personally
like LTO. There are several manafacturers of these now incld. IBM &
Seagate.

For software Veritas NetBackup wins hands down. Legato is clumsy compared
to NetBackup. I like the GUIs much better for NetBackup it is more
complex however.
_____
It all depends on amount of data, backup windows and cost you want to spend
on media, drives and libraries.

We use Veritas Netbackup on NT to backup Solaris, AIX and Tru64 and Windows
server.  We use a 15 slot AIT-1 library but we are moving to AIT-3 15 or 30
slot library.  

This solution will be fine as long as we do not backup more than 250Gb in
one night.  That is the most we can backup over 100Mb/s network.

We find that AIT drives are more cost effective than other solutions we have
looked at.  Then again, I have not looked at SDLT or LTO lately.  The
library is the Spectra 2000 made by spectralogic.  It costs around $11,000
US.
_____
There have been several discussions about ait vs dlt tapes on the
Legato NSR (Networker Save and Restore) list.  Networker is software that
manages backups, a la Veritas.

This is not a Networker or Veritas endorsement, but you can check out the
information on ait and dlt tapes/drives at:

listserv.temple.edu/archives/networker.html

I think you need to subscribe to the list, but you can do that online and
unsubscribe when you're done.
_____

Manufacturer really depends. Many are really good - but it's probably
best, from a support point of view, to stick to your backup server
manufacturer. Here we have a Sun server doing backups, so we bought a Sun
tape library (which Sun OEM from StorageTek). So we get good StorageTek
hardware, but a single source of tech support, which is really convenient.

For the technology, SDLT and LTO seem to be the way of the future. SDLT is
Quantum's technology while LTO is backed by several manufacturers (IBM and
Seagate, among others). I would tend to go the multi-manufacturers way
myself. For your several years worth of DLT, you should keep your current
drive as neither is compatible anyway.
_____

I'm assuming an 8 hour backup window.  Technically 1 SDLT drive would
accomplish this right now.  I have a single SDLT drive and it streams
regularly at 25mb/sec (assuming your clients can keep it going); but
let's pad that a bit.  20mb/sec = 1200mb/min = 72gb/hour = 576gb/8hrs.

However, you may not have enough resource or horsepower to drive the
tape at that speed, so if you pad it with another drive you should be
fine.

As to libraries; I have an ATL P6000, which is probably overkill in your
case but maybe not, depends on growth potential.  ATL makes an M1500
that has 2 SDLT drives and holds 21 tapes.  This is enough capacity for
your needs but may require you changing tapes more often than you would
like.  I would personally look for a library that kept me from changing
tapes at all, only eject/inject for offsite.  The P1000 may do it for
you.

Of course there are other library makers....but I woudl look for
something that could hold at least 4 SDLT drives (or some other uber
drive of your choice) and 400 slots.  If you plan to expand fast I would
say 8 drives and 600 slots.
_____
Just a quick thought, while you are going to get a bunch of replies 
concerning hardware, how many tapes, etc., something else to consider is 
the backup window that you have.  How long is your system going to take 
to backup that data?  Depending on your window, that could impact how 
many drives backing up concurrently, and how fast the drives are (DLT 
slow, but high capacity - AIT drives fast, but expensive and lower 
capacity).
_____
We use AIT tapes, not at AIT-3 (100GB/tape, 36MB/s). The drives
are expensive, but the capacity and speed make up for that, as
does the reliability of the drives. We haven't had a single failure
in over 4 years of use of 4 drives (started with AIT-1).

I've attached a document comparing tape technologies, it's from
www.spectralogic.com. Take it with a grain of salt, it comes from
a company which makes libraries using AIT, but it's conclusions
agree with everything I've heard in surveys, articles and from
talking with vendors.

We use a Qualstar tape library, the price is very good and they
are known to be very reliable. We haven't had any problems with
ours.
_____
we use a tape silo + robot sourced from StorageTek (9710 library, with
500 tape slots). the tapes are DLT700. Monthly full b/u 1.3Tb spread
over 3 sites. Not clear what we would do if we started over from scratch,
AIT looks attractive but we have yet to hear of someone using it in
production.
_____

One word answer sun L180 with veritas netbackup.
_____

If you've already "standardised" on DLT, go for SDLT in 
a smallish jukebox like a STK 9730 if you're looking to 
"improve" on what you've got.  Alternatively, stick with
straight DLT and get extra drives.  However, take into 
account how you will manage the jukebox.  If you're 
looking to use one jukebox with NT backup and Solaris 
backup, you're going to need something to manage which 
software is allowed to use which tapes in which drives.
If you're running NetWorker (since I know it well), you
can set up your jukebox to give all of your bigger 
servers local drives to back up to, and let it manage 
all your backups.

Just got to your second post, since my VPN was down when
I tried to send this one.  I'd guess something like the
9730 would be fine for you:
http://cgi.ebay.co.uk/ws/eBayISAPI.dll?ViewItem&item=2024279913
Alternatively, something from ATL is always a good choice,
and they seem to have more parts on hand when you've got a
problem.

Given your storage needs, I'd recommend one piece of 
network backup software which will handle all of your 
backup requirements.  Legato NetWorker (Solstice Backup) 
and Veritas NetBackup are the two main options in this 
field.  
_____


Joanna Sizemore
Data Integration Manager
Grote Industries, LLC
v (812) 265-8857
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From jason.kruse at teldta.com  Wed May 15 11:35:56 2002
From: jason.kruse at teldta.com (Kruse, Jason K.)
Date: Wed, 15 May 2002 10:35:56 -0500
Subject: SUMMARY: SSH/SFTP lockdown
Message-ID: <5D259859BB8ED2119B720008C75657700AFAABA2@corp0005.teldta.com>

Thanks to:
Daron.Barndon
Davorin Bengez
Greg Gallagher
Michael C. Ibarra
Jim Jones
Dennis Peterson 
Eric Shafto

No real answer that works with Solaris 8/F-Secure.  
Responses included:

Scponly:  http://sublimation.org/scponly/
rksh - Restricts shell nicely, however sftp still can access directories
/bin/true vs /bin/false (no shell access vs no access at all)
OpenSSH w/chroot patch

We're working on converting to OpenSSH, but until that happens we'll just
have to keep our eyes open.

Jason

> -----Original Message-----
> From: Kruse, Jason K. [mailto:jason.kruse at teldta.com]
> Sent: 5/13/2002 8:47 AM
> To: 'sunmanagers at sunmanagers.org'
> Subject: SSH/SFTP lockdown
> 
> 
> We have a customer who has requested using sftp to access files on our
> system.  I would like to restrict their access to their home 
> directory,
> however F-Secure does not provide the chroot ability on Solaris.  
> 
> I attempted to assign the user a bogus shell like /bin/false 
> or noshell but
> sftp just complained.  Other than creating a chroot 
> environment by hand
> (using jail or something similar) does anyone have other 
> suggestions to
> restrict access?
> 
> Jason
> _______________________________________________
> sunmanagers mailing list
> sunmanagers at sunmanagers.org
> http://www.sunmanagers.org/mailman/listinfo/sunmanagers
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From scot.needy at corp.infi.net  Wed May 15 09:20:02 2002
From: scot.needy at corp.infi.net (Scot Needy)
Date: 15 May 2002 09:20:02 -0400
Subject: SUMMARY: Loading perl modules bassed on Solaris uname
In-Reply-To: <1021406764.1227.67.camel@sywn001.engrs.infi.net>
References: <1021406764.1227.67.camel@sywn001.engrs.infi.net>
Message-ID: <1021468824.1261.15.camel@sywn001.engrs.infi.net>

Thanks to those who responded. 

There are a few discussons on the net about this. Apparently the use 
statements are processed before ANY other perl code so... 

use lib $path is equal to use lib "NULL". 
You cannot use a variable yet because perl has not parsed that bit of
code that defines $path. 

The only way I am able to push a dynamic directory onto the INC array is
by wrapping perl in a shell script setting LD_LIBRARY_PATH or 
by adding a bit of ugly code. Still can't seem to get DynaLoader to look
in a different directory so all my .so files are in lib. 

use strict;
use POSIX qw(uname);
my ($uname_s, $uname_r)  = (POSIX::uname())[0,2];

if (($uname_s eq "SunOS") && ($uname_r eq 5.6)) {
      use lib qw(/var/opt/modules/SunOS/5.6
/var/opt/modules/SunOS/5.6/lib);
   } elsif (($uname_s eq "SunOS") && ($uname_r eq 5.7)) {
      use lib qw(/var/opt/modules/SunOS/5.7
/var/opt/modules/SunOS/5.7/lib);
   } elsif (($uname_s eq "SunOS") && ($uname_r eq 5.8)) {
      use lib qw(/var/opt/modules/SunOS/5.8
/var/opt/modules/SunOS/5.8/lib);
   } else {
      die("$0 is only supported on Solaris 2.6, 2.7 & 2.8 \n");
   }

Scot 

On Tue, 2002-05-14 at 16:06, scotrn wrote:
> Hi; 
> 
>  I was wondering if any of you have tried to dynamicly load 
> a module from a directory based on the rev of OS? 
> 
> In other words....
> my ($uname_s, $uname_r) = split('\s',`uname -sr`);
> 
> use lib "/share/local/modules/$uname_s"."/"."$uname_r";
> 
> This seems to work but modules which use DynaLoader to load shared
> objects it's falling back to the original @INC and not looking in 
> my lib path. 
> 
> Trying to support an NFS share which has multiple versions of compuled 
> modules bassed on OS Rev. 
> 
> Thanks 
> Scot 
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From daniel.zhuang at amdocs.com  Wed May 15 15:05:15 2002
From: daniel.zhuang at amdocs.com (Daniel Zhuang)
Date: Wed, 15 May 2002 15:05:15 -0400
Subject: Summary: accidently delete /devices/pseudo
Message-ID: <1607D3EC1FCA2E4F8A078CDC2CFF880B12DDE8@torex1gen.amdocs.com>

Thans to Sal Serafino, Kevin Korb,John Malick, Francisco Puente
Fabrice Guerini, Darren Dunham, Chris Kurtz ,Bolddin Bob, etc

first it doesn't have to reinstall system;
second, it does need to boot into single user mode through cdrom or net.
most experts recommend the way below; some others say copying /dev and /devices from cdrom to /a at single mode. but the "cp" command doesn't copy character devices rightly when I tried.
anyway, the way below is simple and effective .


	Boot from a CD using "boot cdrom -s"
	Mount your root partition under /a
	Rename /a/etc/path_to_inst as /a/etc/path_to_inst_saved
	Do driver reconfigurations on the machine's real root:
		drvconfig -n -r /a
	Touch /a/reconfigure
	Sync and umount /a
	Abort the running kernel using Stop-A
	Reboot using "boot -svra"
		Single user
		Verbose (always nice to see everything
		Reconfigure
		Ask Me
	At the AskMe prompts, take the defaults.  If the system hangs on bootup
	you will at least know what file is giving you trouble and you can try
	to set up a new version of the offending configs using the CDROM.

If you are successful at rebooting the machine into single-user mode, I suggest 
that you "reboot -- -svr" back into single-user to make sure there are no 
problems.  Then, you should be OK to boot into normal mode.




Message: 2
From: Daniel Zhuang <daniel.zhuang at amdocs.com>
To: "'sunmanagers at sunmanagers.org'" <sunmanagers at sunmanagers.org>
Subject: accidently delete /devices/pseudo,help
Date: Wed, 8 May 2002 12:33:23 -0400 

Hi Gurus,
I accidently deleted whole directory of /devices/pseudo on Solaris 2.6 at a 
Ultra5 
when I tried to boot from harddisk, it hungs before the point "Configuring 
devices"
how can I restore ../pseudo whithout rebuild the box

boot -r hungs on the same place

thanks
Daniel Zhuang
Unix Admin, Infrastructure Support Team
--------------------------------------------------------------------------------
--
-------------------------------------------------------------------------------------

The information contained in this message is proprietary of Amdocs,
protected from disclosure, and may be privileged.
The information is intended to be conveyed only to the designated recipient(s)
of the message. If the reader of this message is not the intended recipient,
you are hereby notified that any dissemination, use, distribution or copying of 
this communication is strictly prohibited and may be unlawful. 
If you have received this communication in error, please notify us immediately
by replying to the message and deleting it from your computer.
Thank you.

-------------------------------------------------------------------------------------
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From nickus at mpi-cbg.de  Wed May 15 11:48:30 2002
From: nickus at mpi-cbg.de (Niclas Sodergard)
Date: Wed, 15 May 2002 17:48:30 +0200
Subject: SUMMARY: Transtec 5012 RAID (Zero-D)
In-Reply-To: <200205141117.36254.nickus@mpi-cbg.de>
References: <200205141117.36254.nickus@mpi-cbg.de>
Message-ID: <200205151748.30842.nickus@mpi-cbg.de>

Hi managers,

I solved this problem thanks to Martin Hepworth, Justin Stringfellow, Josh
Glover, Paul Greidanus. Thanks a lot for the quick answers.

It seems like the raid drive was a bit to big. There is a 1TB limit and if I
sliced the raid into two, 1TB, and 500GB then it worked just nicely.

Cheers,
Nickus

> Hi managers
>
> We bought a Transtec 5012 IDE/SCSI raid with 12x160GB Maxtor harddrives (It
> should be the same unit as the Zero-D X-3i - just rebranded by Transtec).
> The unit is configured as a raid 5 with 1 spare disk. The unit works fine
> when connected to either a Windows box or a Linux box. However, when I
> connect it to a Sun Enterprise 220R I run into trouble.
>
> I can see the drive when I use the probe-scsi command in the bootprom. I
> can also see the drive after a drvconfig and then format. But when I select
> the drive in format it cannot autoconfigure it.  I get the following
> output.
>
> I run format with -e flag. It doesn't matter what I specify as the way it
> should autoconfigure, I get the same error all the time.
>
> Does anyone have a good solution?
>
> Thanks,
> Niclas
>
> # output from format -e
>
>       14. c3t0d0 <drive type unknown>
>           /pci at 1f,4000/scsi at 3,1/sd at 0,0
> Specify disk (enter its number): 14
>
>
>
> AVAILABLE DRIVE TYPES:
>         0. Auto configure
>         1. Quantum ProDrive 80S
>         2. Quantum ProDrive 105S
>         3. CDC Wren IV 94171-344
>         4. SUN0104
>         5. SUN0207
>         6. SUN0327
>         7. SUN0340
>         8. SUN0424
>         9. SUN0535
>         10. SUN0669
>         11. SUN1.0G
>         12. SUN1.05
>         13. SUN1.3G
>         14. SUN2.1G
>         15. SUN2.9G
>         16. SUN18G
>         17. other
> Specify disk type (enter its number): 0
> Auto configuration via format.dat[no]?
> Auto configuration via generic SCSI-2[no]? yes
> Mode sense page(3) reports nsect value as 128, adjusting it to -43
> 1. Capacity = -2147482624, with pcyl = -390168 nhead = 128 nsect = -43
> 2. Capacity = -1100566016, with pcyl = 195413 nhead = 128 nsect = -44
> _______________________________________________
> sunmanagers mailing list
> sunmanagers at sunmanagers.org
> http://www.sunmanagers.org/mailman/listinfo/sunmanagers

Cheers,
Nickus

--
Please use the TroubleTicket systems at http://intranet.mpi-cbg.de/tt
when you want to report problems. Please try to avoid direct emails
to the computer department. Thank you.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From barnesdan at yahoo.com  Wed May 15 16:00:08 2002
From: barnesdan at yahoo.com (Dan Barnes)
Date: Wed, 15 May 2002 13:00:08 -0700 (PDT)
Subject: SUMMARY: ndd package
In-Reply-To: <20020515193534.49985.qmail@web11708.mail.yahoo.com>
Message-ID: <20020515200008.53360.qmail@web11708.mail.yahoo.com>

The package is SUNWcsu. I must have somehow deleted
the binary. SUNWcsu cannot be removed of course on a
working system.

Thanks to all.

- db
--- Dan Barnes <barnesdan at yahoo.com> wrote:
> Greetings,
> 
> Does anyone happen to know what SUNW package the ndd
> binary is included in? In minimizing my system I
> must
> inadvertanly took out that package.
> 
> Thanks in advance,
> 
> - db
> LAUNCH - Your Yahoo! Music Experience
> http://launch.yahoo.com
> _______________________________________________
> sunmanagers mailing list
> sunmanagers at sunmanagers.org
>
http://www.sunmanagers.org/mailman/listinfo/sunmanagers
LAUNCH - Your Yahoo! Music Experience
http://launch.yahoo.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From vlader at optonline.net  Wed May 15 16:12:47 2002
From: vlader at optonline.net (Vlade)
Date: Wed, 15 May 2002 16:12:47 -0400
Subject: Summary: FW-1 on Solaris
Message-ID: <088301c1fc4c$e1aab660$9a0a10ac@wknycvr>

I needed a newsgroup or mailing list for checkpoint and I got great
responses.

Thanks all for the quick response.

Most people referred me to the phoneboy site , which I was aware of but never
new there was a mailing list associated:

Alex Shepard,
Moti (thanks)
Steve Mickeler

What I find which I htink will be very helpful is
Http://www.checkpoint.com/services/mailing.html
http://news.checkpoint.com/nntp.html

which was referred to by:
Landi, Laura - CISU-2
Shannon Wimberly
Jerry Kemp
William Hathaway

Anyone know of a specific newgroup for NOKIA IP330-400 running FW-1?

Thanks alot
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From foster at dim.ucsd.edu  Wed May 15 20:19:05 2002
From: foster at dim.ucsd.edu (David Foster)
Date: Wed, 15 May 2002 17:19:05 -0700 (PDT)
Subject: [SUMMARY] DVD: Hardware and Software Solutions
Message-ID: <200205160019.g4G0J5w04779@dim.ucsd.edu>

Software Solutions:
====================

Got quite a range of suggestions, but we decided to go with
Jvrg Schilling's 'cdrecord-ProDVD' solution. Downside is it
is available as binary only (alpha, irix, linux, solaris,
solaris x86, aix, windows), it's not open-source. To make
matters worse, the program requires a key from Jorg and
apparently it can be quite difficult to get in touch with him.
There is a key for Solaris and Linux available in the README
file of the Pro-DVD ftp site.

See the "4 years of DVD recording with cdrecord-ProDVD" link at:

http://www.fokus.gmd.de/research/cc/glone/employees/joerg.schilling/private/c
dre
cord.html

There is a hacked version of cdrecord which has the DVD stuff
added in (apparently fairly simple since much of the DVD features
have been there for years). Jorg does not support this version.
If anyone would like a pointer to this version email me.

Hardware Solutions:
====================

Most recommended was the Pioneer A03/A04, which is what the
cdrecord developers use. It is EIDE and supports CD-R/CD-RW.
I found the 305S/505S models, which are the newer SCSI models
and are a bit faster and also support CD-R/CD-RW.

YoungMinds was suggested, but it is overkill for what we need:
   http://www.youngmindsinc.com

SmartDax was recommended, it's a jukebox and again much more
than we need:
   http://www.smartdax.com


Thanks to:

Thomas Anders
VICTOR KARPOVICH
B. Srinivas

>
> We are in urgent need of hardware and software solutions
> for implementing the archival of very large data-sets to DVD.
>
> The CDrecord package from Joerg Schilling has some vague notes
> about DVD support in his "cdrecord PRO-DVD" program, but the
> notes on its download page state:
>
>         The binaries are intended for testing purposes. They allow
>         you to write either complete DVD-R & DVD-RW media when using
>         the -dummy option or to write up to 1 GB of real data
>         to a single media.
>
> This hardly seems useful.
>
> Can anyone make hardware and software (free or commercial)
> recommendations for implementing DVD under Solaris?
>
> I'm familiar with the recent announcement about a new "Blu-Ray" DVD
> standard, but this won't be available until next year, and we cannot
> wait that long. For details see:
>
> 	http://www.newscientist.com/news/news.jsp?id=ns99991952
>
> Thanks!

  =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
   David Foster    National Center for Microscopy and Imaging Research
    Programmer/Analyst     University of California, San Diego
    dfoster at ucsd.edu       Department of Neuroscience, Mail 0608
    (858) 534-7968         http://ncmir.ucsd.edu/
  =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=

   "The reasonable man adapts himself to the world; the unreasonable one
   persists in trying to adapt the world to himself.  Therefore, all progress
   depends on the unreasonable."   -- George Bernard Shaw
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From akrasanth at velankani.com  Thu May 16 04:01:37 2002
From: akrasanth at velankani.com (AK Rasanth)
Date: Thu, 16 May 2002 13:31:37 +0530
Subject: SUMMARY:port problem
Message-ID: <0205161331375G.01176@rasanth>

Hi,
Thanks to all those who gave a solution .the solution is to use lsof  which 
can be downloaded from 
www.sunfreeware.com
Thanks&Regards,
A.K.Rasanth
ph: 91-80-8522507 Ext 1626

--------------------------------------------------------------
Velankani Information Systems Ltd, Bangalore, India
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From aredondo at indra.es  Thu May 16 07:14:09 2002
From: aredondo at indra.es (Redondo Escobar, Alejandro)
Date: Thu, 16 May 2002 13:14:09 +0200
Subject: SUMMARY: Which package to install the C pre-processor on
  Solaris8
Message-ID: <1292796112760E419700AC368B722D8A63AF71@MADARRCLEX3.indra.es>

Thanks everybody!
The key: grep /usr/ccs/lib/cpp /var/sadm/install/contents
The keyword: SUNWsprot


-----Mensaje original-----
De: Redondo Escobar, Alejandro
Enviado el: jueves, 16 de mayo de 2002 11:30
Para: sunmanagers at sunmanagers.org
Asunto: Which package to install the C pre-processor on Solaris8


Hi all sunmanagers!
I am trying to lauch an aplication who it's needing the /usr/ccs/lib/cpp "C
language pre-processor", and this is not installed on the host.

Please, does somebody knows, what's the name of the package (from the
Solaris8
cdroms o whatever) to "pkgadd" it??
I have the preprocessor installed on another solaris8 host, but unfortunately
I get no hits from "pkginfo -l |grep -i cpp".

Thanks for all in advance!


-----------------------------------------------------------------
Este correo electrsnico y, en su caso, cualquier fichero anexo al mismo,
contiene informacisn de caracter confidencial exclusivamente dirigida a su
destinatario o destinatarios. Queda prohibida su divulgacisn, copia o
distribucisn a terceros sin la previa autorizacisn escrita de Indra. En el
caso de haber recibido este correo electrsnico por error, se ruega
notifmquese
inmediatamente esta circunstancia mediante reenvmo a la direccisn electrsnica
del remitente.
-----------------------------------------------------------------
The information in this e-mail and in any attachments is confidential and
solely for the attention and use of the named addressee(s). You are hereby
notified that any dissemination, distribution or copy of this communication
is
prohibited without the prior written consent of Indra. If you have received
this communication in error, please, notify the sender by reply e-mail.
-----------------------------------------------------------------
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


-----------------------------------------------------------------
Este correo electronico y, en su caso, cualquier fichero anexo al mismo,
contiene informacion de caracter confidencial exclusivamente dirigida a su
destinatario o destinatarios. Queda prohibida su divulgacion, copia o
distribucion a terceros sin la previa autorizacion escrita de Indra. En el
caso de haber recibido este correo electronico por error, se ruega notifiquese
inmediatamente esta circunstancia mediante reenvio a la direccion electronica
del remitente.
-----------------------------------------------------------------
The information in this e-mail and in any attachments is confidential and
solely for the attention and use of the named addressee(s). You are hereby
notified that any dissemination, distribution or copy of this communication is
prohibited without the prior written consent of Indra. If you have received
this communication in error, please, notify the sender by reply e-mail.
-----------------------------------------------------------------


-----------------------------------------------------------------
Este correo electrsnico y, en su caso, cualquier fichero anexo al mismo,
contiene informacisn de caracter confidencial exclusivamente dirigida a su
destinatario o destinatarios. Queda prohibida su divulgacisn, copia o
distribucisn a terceros sin la previa autorizacisn escrita de Indra. En el
caso de haber recibido este correo electrsnico por error, se ruega notifmquese
inmediatamente esta circunstancia mediante reenvmo a la direccisn electrsnica
del remitente.
-----------------------------------------------------------------
The information in this e-mail and in any attachments is confidential and
solely for the attention and use of the named addressee(s). You are hereby
notified that any dissemination, distribution or copy of this communication is
prohibited without the prior written consent of Indra. If you have received
this communication in error, please, notify the sender by reply e-mail.
-----------------------------------------------------------------
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From Richard.Warden at natwest.com  Thu May 16 07:33:04 2002
From: Richard.Warden at natwest.com (WARDEN, Richard, IT-NWS)
Date: Thu, 16 May 2002 12:33:04 +0100
Subject: SUMMARY:"sudo" on Solaris 5.4
Message-ID: <7FE970CBA965D2119E5B0008C70D8A41029420FB@EXGEMS00202>

Thanks to all who replied to my requests.  Most people pointed me in the
direction of www.sunfreeware.com <http://www.sunfreeware.com> , however,
this site only contains binaries for Solaris 2.5.1 onwards.

Special thanks though, to Mark Hargrave, who e-mailed a copy of the "sudo"
command which seems to work fine.


Regards,
--
Richard Warden
UNIX / Senior SWIFT Systems Administrator
Systems Support Analyst Team (NatWest Stockbrokers)
Coutts Group IT 
020 7895 5267
ITS: 76508 5267
Richard.Warden at NatWest.com <mailto:Richard.Warden at NatWest.com> 



Issued by NatWest Stockbrokers Limited, a member of the London Stock
Exchange and regulated by the FSA. Registered Office : 55 Mansell Street,
London E1 8AN. Registered number 1959479 England

--------------------------------------------------------------------------------------------
"National Westminster Bank plc is registered in England No 929027. Registered Office: 135 Bishopsgate, London EC2M 3UR.
National Westminster Bank plc is regulated by IMRO, SFA and Personal Investment Authority.  A member of the NatWest and Gartmore Marketing Group advising on the life assurance, pensions and unit trust products only of that Marketing Group.

This e-mail message is confidential and for use by the addressee only.  If the message is received by anyone other than the addressee, please return the message to the sender by replying to it and then delete the message from your computer.
Internet e-mails are not necessarily secure.  National Westminster Bank plc does not accept responsibility for changes made to this message after it was sent."
--------------------------------------------------------------------------------------------
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From chris at ludwig-alpha.unil.ch  Thu May 16 07:41:51 2002
From: chris at ludwig-alpha.unil.ch (Christian Iseli)
Date: Thu, 16 May 2002 13:41:51 +0200
Subject: SUMMARY: Will device names change when adding a D1000 to an
  UE450 ? 
Message-ID: <200205161141.NAA10940@ludwig-alpha.unil.ch>

Hi,

Thanks to Matthew Stier, Darren Dunham, Jay Lessert, and Kris Briscoe for 
their answers.

Short answer (and consensus) is: no.


A while ago, I asked:

> I will soon need to add a dual port SCSI board with a D1000 storage device
> to a Sun Ultra Enterprise 450, running Solaris 2.6 with SDS.
>
> The machine currently has 20 internal disk slots (18 full) and two external
> D1000 devices, each with 12 disks.  The PCI slot layout is as follows:
> 10: empty
>  9: empty
>  8: empty
>  7: Sun Gigabit ethernet board
>  6: dual port SCSI board connected to external D1000
>  5: dual port SCSI board connected to internal disk rack
>  4: dual port SCSI board connected to external D1000
>  3: empty
>  2: empty
>  1: dual port SCSI board connected to internal disk rack
>
> The dual port SCSI boards are all Symbios 53c875.
>
> When I add in the new PCI-SCSI board, should I worry that it might change the
> current naming of the disks, and if so, what's the best way to deal with it ?

In Matthew's words:
SunOS keeps a flat file database of devices, mapping their OBP path, to an
OS instance.  The file is /etc/path_to_inst, and unless you remove the instances
for existing devices, you won't run into any conflicts.

> Any words of advice on which slot to use for the new board ?

None received.

Regards,
					Christian
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From chris at ludwig-alpha.unil.ch  Thu May 16 08:03:50 2002
From: chris at ludwig-alpha.unil.ch (Christian Iseli)
Date: Thu, 16 May 2002 14:03:50 +0200
Subject: SUMMARY:Solaris 8 on UltraEnterprise 450 install
Message-ID: <200205161203.OAA11009@ludwig-alpha.unil.ch>

Hi,

Thanks to Simon Kuhn, Jay Lessert, Adam Levin, Krister Bergman, Thomas Anders, 
and Karl Vogel for their answers.

Adam thought that maybe I was pushing it a bit, having so many disks attached 
to the poor beast.  Since it's all we're gonna get for a while, there's no 
alternative.  He pointed me to a page
http://www.westnet.com/~levins/e450scsi.html
describing how the PCI busses and SCSI controlers are assigned in a 450.  The 
least I could do was try to balance the load among the several PCI controlers.

Karl provided some pointers to reference material (white papers) about solaris 
tuning, and some more meterial pasted below:
---
http://sunsolve.sun.com/private-cgi/retrieve.pl?type=2&doc=stb/1442
White Papers/Tech Bulletins 1442
Delivering Performance on Sun: System Tuning
Greg Schmitz and Allan Esclamado
30-Apr-1999

This document focuses on techniques for performance tuning for the Sun
computing environment.  It is aimed at system administrators.  Each chapter
concentrates on a different subsystem of the computing environment (e.g.,
Tuning the Solaris Kernel, Memory, Tuning Disk Subsystems, etc.) and the
specific things that can be done to increase performance.

-------------------------------------------------------------------------
If you have an application that's an incredible swap hog, or the system is
really slowing down, try adding the lines below to /etc/system and
rebooting.  I run with these settings and they've never caused me trouble.

*
* Swap
*   System keeps 1/8th of all memory for swap, which is too much for
*   a 4GB system.  Reduce that to 32 Mbytes (4096 8K pages).
set swapfs_minfree=4096
*
* Memory management
*
*   http://www.carumba.com/talk/random/tuning-solaris-checkpoint.txt
*   Tuning Solaris for FireWall-1
*   Rob Thomas robt at cymru.com
*   14 Aug 2000
*
*   On firewalls, it is not at all uncommon to have quite a bit of
*   physical memory.  However, as the amount of physical memory is
*   increased, the amount of time the kernel spends managing that
*   memory also increases.  During periods of high load, this may
*   decrease throughput.
*
*   To decrease the amount of memory fsflush scans during any scan
*   interval, we must modify the kernel variable autoup.  The default
*   is 30.  For firewalls with 128MB of RAM or more, increase this
*   value.  The end result is less time spent managing buffers,
*   and more time spent servicing packets.
set autoup = 120

-------------------------------------------------------------------------
Check the size of your inode caches, which keep track of previously-
accessed files.  Run the DNLC script below as root to see your hit-rate
percentage.  If it's under 90-95%, you need to up the cache size.
The easiest way to do that is change maxusers in /etc/system to a nice
high number like 2048.

Run "mount" to see how your filesystems are set up.  You can mount
filesystems with "noatime" and journalling ("logging") turned on,
meaning don't bother updating the access time whenever a file is opened.
We use this under Solaris-8, and it makes a *huge* difference if you're
doing something to a large number of small or medium sized files.

---------------------------------------------------------------------------
#!/bin/sh
#
# NAME:
#    dnlc
#
# SYNOPSIS:
#    dnlc
#
# DESCRIPTION:
#    "dnlc" reports on Directory name lookup cache statistics from
#    the kernel.  This corrects a bug in vmstat.
#
#    To change the kernel values, add something like this
#    to /etc/system and reboot.  Both "nnn" numbers should be the
#    same.
#
#         set ncsize = nnn
#         set ufs_ninode = nnn
#
# AUTHOR:
#    Kimberley Brown - UKAC Kernel Support
#    comp.unix.solaris

PATH=/bin:/usr/bin:/usr/local/bin
export PATH

adb -k /dev/ksyms /dev/mem <<END
="**  Directory/Inode Cache Statistics  **"
="----------------------------------------"
ufs_ninode/D"Inode cache size"
ncsize/D"Directory name cache size"
ncstats/D"# of cache hits that we used"
+/D"# of misses"
+/D"# of enters done"
+/D"# of enters tried when already cached"
+/D"# of long names tried to enter"
+/D"# of long name tried to look up"
+/D"# of times LRU list was empty"
+/D"# of purges of cache"
*ncstats%1000>a
*(ncstats+4)%1000>b
*(ncstats+14)%1000>c
<a+<b+<c>n
<a*0t100%<n=D"Hit rate percentage"
="(See /usr/include/sys/dnlc.h for more information)"
END

exit 0
---

A while ago, I said:
> I will soon be upgrading an UltraEnterprise 450 machine to Solaris 8 2/
> 02, and  I still have a few nagging doubts...

> The machine will have 4 GB RAM, about 1 TB disks, and Gigabit ethernet
> (Sun Vector Gigabit Ethernet PCI NIC).  It is used mainly as an NFS
> file server, and as a web server.  The main data partition will be a
> SDS concat/stripe of about 900 GB.  The disks are in the built-in
> racks, and in three D1000 cabinets.

> doubt #1: is the vge driver included in the Solaris 8 2/02 release,
> and part  of the normal install ?

Krister warned me that my vge board was not supported in Solaris 8, and was 
quite right about it.  I subsequently opened a call (the machine has a Silver 
support contract) and Sun Switzerland was nice enough to send me a replacement 
GE board for free.

> doubts #2: I have read through the "Solaris Tunable Parameters
> Reference  Manual", and came to the part on Per-Route Metrics, which
> explains that for  gigabit ethernet devices, tcp_recv_hiwat should be
> increased.  Question is:  how much ?  Also, should tcp_xmit_hiwat be
> increased as well ?

No specific answer.

> doubt #3: Am I correct in assuming that tcp_max_buf should be set
> around  104,857,600 (i.e., 100 times its default value) ?

No specific answer.

> doubt #4: the machine currently has 1.7 GB RAM and Solaris 2.6 (+
> patches).   If I look at vmstat, the sr column appears to stay at 0.
> Should I then set  dopageflush to 0 ?  Or am I better off increasing
> autoup ?

No specific answer.

> doubt #5: it sometimes happens in the current setting that a run-away
> perl script from the web server eats up all the swap space by filling
> up /tmp (mounted from swapfs) and the system pretty much hangs at that
> point.  Is setting tmpfs:tmpfs_minfree to some megabytes the right way
> to prevent the  freeze and allow root logins to kill the bad process ?
>  Is the default value  really 256 *bytes* ?

Jay had a simple solution:
Keep it simple.  Just limit /tmp size in vfstab:
    swap        -       /tmp    tmpfs   -       yes     size=500m

> doubt #6: what else am I missing ? ;) 

Simon said to be a bit weary of the 02/02 release, as the statd daemon has a 
known bug that will cause it to fail every now and then.  I can confirm this 
is the case, and it has to be restarted about once daily.  There's supposed to 
be a patch available in mid-May.

Regards,
					Christian
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From chris at ludwig-alpha.unil.ch  Thu May 16 08:20:46 2002
From: chris at ludwig-alpha.unil.ch (Christian Iseli)
Date: Thu, 16 May 2002 14:20:46 +0200
Subject: SUMMARY: can't automount a newly exported partition in /net 
Message-ID: <200205161220.OAA11106@ludwig-alpha.unil.ch>

Hi,

Thanks to Andrew Rotramel, John Riddoch, Ted John, David Harrington, 
Darren Dunham, Umesh Potbhare, Matthew Stier, and Rik Schneider for their 
answers.

Unfortunately, it seems there is no way to get this to work, short of either 
stopping all the processes that use some exports of server A, or of rebooting 
client B.

Note that Solaris 8 clients have not this problem.  For them it just works.

Here is some advice I received.
> Try one of

> /etc/init.d/nscd stop
> /etc/init.d/nscd start
> or
> automount
---
> I think it has something to do with when the client b queries server A for
> what is share on /net. This happens at boot but I have not been able to find
> out how to force it. I had tried restarting many servers including nfs and
> auotmounter.
---
> Don't use /net.  Set up a nice automount map for your servers.

> As you've seen /net has limitations.  Number one is that it only scans
> when the server is mounted.  So if you can reboot, I'm sure you can
> kill all the processes that use it.  Once they stop, you can unmount
> everything under /net/host.  The next time it mounts, it will rescan.

> All that pain is gone if you set up your own map. 

I can see the point, but on the other hand, /net is so convenient.  Also, 
Solaris 8 seems to get it right...
---
> Use /etc/init.d/nfs.client stop and /etc/init.d/nfs.client start, Run
> it on Client B as a root. Even if after this if it donot see Wait and
> it should see that after nfs timeout. 
---
> Try running "/usr/sbin/automount" as root. 
---
> Try running automount by hand on the client machine. 
---

I asked:
> I have a bit of a NFS problem.  I have a server A, and a client B.
> server A  is running Solaris 8, and client B is running Solaris 2.6.

> Server A exports: /export/old

> and client B accesses it as /net/A/export/old

> Now, I add a new filesystem to A as /export/new and share it as
> appropriate

> but when I try to access from client B /net/A/export/new I get: /net/A/
> export/new: No such file or directory

> However, I have no problem mounting it through the mount command.

> Is there any way, short of rebooting, to signal to B that /net/A/
> export/new is  actually available ?  I tried "/etc/init.d/autofs stop/
> start", but there are  processes on B that continuously use /net/A/
> export/old, and stopping them is a  pain... 
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From craigg at psu.edu  Thu May 16 09:21:17 2002
From: craigg at psu.edu (Craig L. Gruneberg)
Date: Thu, 16 May 2002 09:21:17 -0400
Subject: SUMMARY II- UFSDUMP and single files
Message-ID: <3CE3B24D.63BADC67@psu.edu>

After much help from the group (Thanks all!) I seem to have narrowed
this down to a file size problem. When I trussed the following command:

# truss ufsdump 0f /dev/rmt/2 /zoso/pc_image/C.diskimage | tee

I got:

llseek(3, 0x000000070AE94000, SEEK_SET)         = 0x70AE94000
read(3, "\0\0\0\0\0\t02 U <E2B8 g".., 8192)     = 8192
llseek(3, 0x000000070DA06000, SEEK_SET)         = 0x70DA06000
read(3, "\0\0\0\0\0\t02 U <E2B8 g".., 8192)     = 8192
  DUMP: Mapping (Pass I) [regular files]
write(2, "     D U M P :   M a p p".., 41)      = 41
stat64("/zoso/pc_image/C.diskimage", 0xFFBEE0E8) = 0
stat("/zoso/pc_image/C.diskimage", 0xFFBEDBF0)  Err#79 EOVERFLOW
  DUMP: Cannot find filesystem mount point for
`/zoso/pc_image/C.diskimage'
write(2, "     D U M P :   C a n n".., 76)      = 76
getuid()                                        = 0 [0]
seteuid(0)                                      = 0
  DUMP: The ENTIRE dump is aborted.
write(2, "     D U M P :   T h e  ".., 36)      = 36
llseek(0, 0, SEEK_CUR)                          = 83209
_exit(3)

which I assume points to the file being too big. I sized my image on
this go round to 10GB so as to not outgrow it and apparently ufsdump, in
spite of being declared "large file aware' in the largefile man page,
really doesn't like a file this size.

I did find that you can compress the SUNPCi image quite nicely (careful-
you need a updated version of gzip to handle a file this size and unlike
compress, gzip wipes the original leaving only the compressed version).
My 10GB image, which is only a 2.8GB C drive in Win2k, compresses down
to 1.5GB.

-craig
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From joaquin.domenech at dutchtone.nl  Thu May 16 09:25:18 2002
From: joaquin.domenech at dutchtone.nl (joaquin.domenech at dutchtone.nl)
Date: Thu, 16 May 2002 15:25:18 +0200
Subject: [SUMMARY]  (O-T)  reverse proxy
Message-ID: <3F24D8FD4C45D41183CE0008C71E67A503A5708E@dtexc03.dutchtone.nl>

Hi,

regarding the recommendations about reverse proxies I asked for, these are
the results:

RP on dedicated machines
cacheflow
NetApp's NetCache
Cisco's CacheEngine

sofware (commercial):
Inktomi - (someone pointed that is very expensive)
Iplanet Proxy Server

freeware:

OOPS (search Freshmeat.net for an URL)
Apache
Squid


many thanks to:

Edwin Groenescheij
Hendrik Visage
Ray McCaffity
John Riddoch
Dave Floyd



===========================================================
De verzonden informatie is uitsluitend bestemd voor de geadresseerde
natuurlijke persoon of rechtspersoon en bevat mogelijk vertrouwelijke en/of
geprivilegeerde gegevens. Met uitzondering van de geadresseerde persoon is
het niet toegestaan de informatie openbaar te maken, te kopikren, te
verspreiden of anderszins actie te ondernemen op basis van de informatie.
Indien u de informatie abusievelijk heeft ontvangen, neem dan contact op met
de afzender en verwijder de informatie uit alle computers. Dutchtone staat
niet in voor de juiste en complete verzending van de informatie, noch is zij
aansprakelijk voor de vertraagde ontvangst hiervan.

The information transmitted is intended exclusively for the person or entity
to which it is addressed and may contain confidential and/or privileged
material. Any disclosure, copying, distribution or other action  based upon
the information by persons or entities other than the intended recipient is
prohibited. If you receive this information in error, please contact the
sender and delete the material from any and all computers. Dutchtone does
not warrant a proper and complete transmission of this information, nor does
it accept liability for any delays.
===========================================================
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From scotrn at cox.net  Thu May 16 09:30:54 2002
From: scotrn at cox.net (scotrn)
Date: 16 May 2002 09:30:54 -0400
Subject: SUMMARY: SUMMARY: Loading perl modules bassed on Solaris uname
In-Reply-To: <1021470821.1263.23.camel@sywn001.engrs.infi.net>
References: <1021406764.1227.67.camel@sywn001.engrs.infi.net> 
  <1021470821.1263.23.camel@sywn001.engrs.infi.net>
Message-ID: <1021555855.1514.5.camel@sywn001.engrs.infi.net>

And the winner is 'BEGIN' thanks for the feedback. 
This code will add a search path based on OS Rev. 
BEGIN {
   use POSIX qw(uname);
   my ($uname_s, $uname_r)  = (POSIX::uname())[0,2];
   unshift(@INC, "/var/opt/modules/$uname_s/$uname_r" );
}

If you wish to follow the rest of this problem I have been asked to 
take it off this list so I am now posting to beginners at perl.org
	Thread : DynaLoader and lib problems. 

The above code works unless your module uses DynaLoader to load a
sharred obj. For some reason it isn't searching 
$dir/$archname/auto as documented in `perldoc lib`

Module I am using is Solaris::Kstat

Scot 

 

On Wed, 2002-05-15 at 09:53, scotrn wrote:
> Thanks to those who responded. 
> 
> There are a few discussons on the net about this. Apparently the use 
> statements are processed before ANY other perl code so... 
> 
> use lib $path is equal to use lib "NULL". 
> You cannot use a variable yet because perl has not parsed that bit of
> code that defines $path. 
> 
> The only way I am able to push a dynamic directory onto the INC array is
> by wrapping perl in a shell script setting LD_LIBRARY_PATH or 
> by adding a bit of ugly code. Still can't seem to get DynaLoader to look
> in a different directory so I created a sun4-solaris/auto directory under 
> each 5.6,5.7 amd 5.8 directory to hold my .so files  
> 
> use strict;
> use POSIX qw(uname);
> my ($uname_s, $uname_r)  = (POSIX::uname())[0,2];
> 
> if (($uname_s eq "SunOS") && ($uname_r eq 5.6)) {
>       use lib qw(/var/opt/modules/SunOS/5.6);
>    } elsif (($uname_s eq "SunOS") && ($uname_r eq 5.7)) {
>       use lib qw(/var/opt/modules/SunOS/5.7);
>    } elsif (($uname_s eq "SunOS") && ($uname_r eq 5.8)) {
>       use lib qw(/var/opt/modules/SunOS/5.8);
>    } else {
>       die("$0 is only supported on Solaris 2.6, 2.7 & 2.8 \n");
>    }
> 
> Scot 
> 
> On Tue, 2002-05-14 at 16:06, scotrn wrote:
> > Hi; 
> > 
> >  I was wondering if any of you have tried to dynamicly load 
> > a module from a directory based on the rev of OS? 
> > 
> > In other words....
> > my ($uname_s, $uname_r) = split('\s',`uname -sr`);
> > 
> > use lib "/share/local/modules/$uname_s"."/"."$uname_r";
> > 
> > This seems to work but modules which use DynaLoader to load shared
> > objects it's falling back to the original @INC and not looking in 
> > my lib path. 
> > 
> > Trying to support an NFS share which has multiple versions of compuled 
> > modules bassed on OS Rev. 
> > 
> > Thanks 
> > Scot 
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From gio at uslink.net  Thu May 16 09:43:52 2002
From: gio at uslink.net (Giovanni Navarrette)
Date: Thu, 16 May 2002 08:43:52 -0500 (CDT)
Subject: SUMMARY: Best way to move lots of data to new system?
Message-ID: <Pine.GSO.4.05.10205160829050.24517-100000@netapp>

Hey everyone:

Sorry for the delay in getting this summary out..., been a horribly busy
couple of weeks. THANKS for the tons of responses y'all sent..., here's
basically the options I got:

---
rsync: Offers compression over the network, you can do it as a 'one big
sync during a maintenance window and update incrementally via cronjobs'

NFS mount/cp: no extra programs needed..., just nfsmount the new RAID and
copy the data over...., rough guesses for copy time were about an hour for
20GB. The most common method was: 'tar/compress/copyormove/uncompress/untar'

dd over network: 'dd if=/dev/rdsk/cXtXdXsX bs= YYk| rsh yourmachine dd 
of=/dev/rdsk/cXtXdXsX' where YY should be the MTU of your network(k was
mentioned for Kilobyte...)

ufsdump over network: ufsdump 0f - /dev/rdsk/c0t3d0s7 | rsh yourmachine 
'(cd /home; ufsrestore rf -)'

tar/ssh over network: cd /LocalParentDir, ssh -x OldHost "cd
/RemoteParentDir ; tar -cf - DataDir" | tar -xpvf -

directly connect new RAID to new old system/old raid: then just cp

ftp: only problem is permissions :\
---

Right now the rsync idea is lookin' pretty good. I still have to get the
RAID up and the server up to production level, so it will still be a
couple of weeks before I try this. I'll let y'all know what happens :)

Thanks for everyone's help! Y'all are the best!! Have a great day!!!

-------------------------------------
Giovanni Navarrette
USLink Internet Systems Administrator
e-Mail: gio at uslink.net
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From martha.crocker at philips.com  Thu May 16 12:33:40 2002
From: martha.crocker at philips.com (martha.crocker at philips.com)
Date: Thu, 16 May 2002 11:33:40 -0500
Subject: SUMMARY: Sendmail - Where is relay host defined? OR Ah Ha!!
  Sendmail relayhost mystery solved! 
Message-ID: <OFF45D6D90.42AEC1C7-ON86256BBB.005A51FB@diamond.philips.com>

Many, many. many thanks to all who responded and especially to:

Morten Jxrgensen    &
Fabrice Guerini

who made me look at that fact that at some point, I mounted /var/mail from
this server it was
trying to make the mailhost.  Problem solved.  Huge sigh of relief and now
I must go through
the office and explain what that quite loud HA! was when the problem was
solved.  You are
a sanity-saving group!

Thankyou, thankyou, thankyou everyone!

Martha Crocker
UNIX System Administrator
Philips Semiconductors
9651 Westover Hills Blvd.
M/S 406
San Antonio, Texas 78251
(210)522-7468  voice
(210)522-7300  fax
martha.crocker at philips.com

The information contained in this message is confidential and may be
legally privileged. The message is intended solely for the addressee(s).
If you are not the intended recipient, you are hereby notified that any
use, dissemination, or reproduction is strictly prohibited and may be
unlawful. If you are not the intended recipient, please contact the sender
by return e-mail and destroy all copies of the original message.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From martha.crocker at philips.com  Thu May 16 13:20:38 2002
From: martha.crocker at philips.com (martha.crocker at philips.com)
Date: Thu, 16 May 2002 12:20:38 -0500
Subject: SUMMARY #2: Sendmail - Where is relay host defined? OR Ah Ha!!
  Sendmail relayhost mystery solved! 
Message-ID: <OFA1275260.97363A5F-ON86256BBB.005EB941@diamond.philips.com>

It was brought to my attention that in my euphoric state after finally 
solving the problem, that my SUMMARY
didn't really contain enough info about either the original problem or the 
solution.  Sorry, here we go again:

Solution:

If /var/mail is nfs mounted from an nfs server, sendmail, at startup, will 
assign the ${ms} variable to the nfs
server.  In my case, I had done this.  Not that I intended to, but 
somehow, I had done this on this one machine.
So, all I had to do was get rid of that and have /var/mail be local again 
and restart sendmail. 



Original Problem:

Good morning all,

I have a problem that is driving me nuts.  Out of my 80 machines, I have 
one that refuses to send mail.  Actually, it doesn't
refuse to send mail, it refuses to find the correct relay, so the machine 
it insists on sending mail through won't allow it to, so
mail doesn't go out.  I'm at my wit's end.  I have two identical machines 
- same Sendmail version, same OS, same patch levels,
same sendmail.cf, same DNS and NIS servers, same nsswitch.conf.  One sends 
mail to the correct relay and one doesn't.
I have mailhost and relayhost defined in DNS.  I just can't figure out 
where this machine is getting it's relay information from.
It is an E3500, Solaris 8 with recommended and security patch cluster from 
April 2002, Sendmail 8.11.6
There must be somewhere where this machine has had hardcoded the relay, 
but I can't find it for the life of me. 

When I look at the syslog messages on both machines, the only difference 
is the "relay=" part.  The machine that sends
mail out says "relay=mailhost.xxx.xx.xxxx.com" and the machine that 
doesn't, says "relay=hostname [IP address] "  where
hostname is the wrong mailhost and the ip address is in brackets.  I'm 
stumped.

Anyone have any ideas where this machine is getting it's incorrect relay 
information from?

Additional information added in another post:

I have received many, many  very helpful replies so far, but let me 
clarify a couple of things.   I've also tracked down some
more information.  I'm trying to wade through Sendmail 2nd Edition to find 
help. 

1)  the sendmail.cf files on the machines are identical - I ftp'd the one 
from the working host. 
2)  The DS entry is there and correct, although I've tried some of your 
suggestions just to see.
        I've even put the name mailhost.xx.xxx.xxx.com in there instead of 
the $m stuff to see
        if I could force it.  It's not getting the information from that 
line.  It's getting it from somewhere
        else.
3)  /etc/hosts does not have mailhost defined on either machine (tried 
that before, too)
4)  I can ping the mailhost using the hostname, the fully qualified 
hostname, mailhost and relayhost
5)  When I do /usr/lib/sendmail -d0.1 -bt < /dev/null   I get this:

On the working host:

Version 8.11.6+Sun
 Compiled with: LDAPMAP MAP_REGEX LOG MATCHGECOS MIME7TO8 MIME8TO7
                NAMED_BIND NDBM NETINET NETINET6 NETUNIX NEWDB NIS NISPLUS
                QUEUE SCANF SMTP USERDB XDEBUG

============ SYSTEM IDENTITY (after readcf) ============
      (short domain name) $w = host1
  (canonical domain name) $j = host1.xxx.xx.xxxxxxxx.com
         (subdomain name) $m = xxx.xx.xxxxxxxx.com
              (node name) $k = host1
========================================================

BUT on the non-working host, I get an extra variable defined:

Version 8.11.6+Sun
 Compiled with: LDAPMAP MAP_REGEX LOG MATCHGECOS MIME7TO8 MIME8TO7
                NAMED_BIND NDBM NETINET NETINET6 NETUNIX NEWDB NIS NISPLUS
                QUEUE SCANF SMTP USERDB XDEBUG

============ SYSTEM IDENTITY (after readcf) ============
      (short domain name) $w = host2
  (canonical domain name) $j = host2.xxx.xx.xxxxxxxx.com
         (subdomain name) $m = xxx.xx.xxxxxxxx.com
              (node name) $k = host2
             (mailserver) ${ms} = wizard
========================================================

This is where I'm stumped - where is that ${ms} defined?  Why is it 
defined on this system and not the other?
I installed both these machines from the same CD and thought I made all 
the same choices when it came to
installation options.  I've never entered anything in for defining a 
mailserver that I can recall. 

I thank everyone who has so far taken the time to try to help me and thank 
in advance anyone who can answer
this puzzle! 

Martha Crocker
UNIX System Administrator
Philips Semiconductors
9651 Westover Hills Blvd.
M/S 406
San Antonio, Texas 78251
(210)522-7468  voice
(210)522-7300  fax
martha.crocker at philips.com

The information contained in this message is confidential and may be 
legally privileged. The message is intended solely for the addressee(s). 
If you are not the intended recipient, you are hereby notified that any 
use, dissemination, or reproduction is strictly prohibited and may be 
unlawful. If you are not the intended recipient, please contact the sender 
by return e-mail and destroy all copies of the original message.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From thesunlover2002 at yahoo.com  Thu May 16 13:25:34 2002
From: thesunlover2002 at yahoo.com (John Lee)
Date: Thu, 16 May 2002 10:25:34 -0700 (PDT)
Subject: SUMMARY: ndd Sun Quad interface to "100Mbps & half-duplex"
Message-ID: <20020516172534.35739.qmail@web14906.mail.yahoo.com>

Question:
I now want to use 'ndd' to make one of the interfaces
on this Sun Quad card use "100 Mbps & half-duplex".
The system is running Solaris 2.6. Here is what I
think to do:
# ndd -set /dev/qfe instance 3
# ndd -set /dev/qfe adv_autoneg_cap 0
# ndd -set /dev/qfe adv_100fdx_cap 0
# ndd -set /dev/qfe adv_100hdx_cap 1
# ndd -set /dev/qfe adv_10fdx_cap 0 (optional)
# ndd -set /dev/qfe adv_10hdx_cap 0 (optional)
Because this is a critical live server, I am kind of
nervous to do it without 100% sure. Please help me to
know if the above method is 100% correct.

Answers:
1) One thing you want to check is your switch or hub
settings for that port. You can do this on the fly but
the main thing is to make sure that the port that you
are connected to on the switch is set to what you want
and that auto_negotiation is disabled on the switch.
(Thanks Debbie)

2) That looks pretty good, but you'll want to move the
'adv_autoneg_cap' line to the end. When that line is
hit, it will reinitialize the interface. Your
interface will probably go offline for 3 - 4 seconds
when it does so. (This is a good point, and I will
post another question to discuss this). (Thanks Darren
Dunham)

3) 'ndd -set /dev/qfe adv_autoneg_cap 0' is the last
thing on the ndd commands - this is the one that
actually 'forces' the interface to reset its speed.
Also why half duplex? This means you'll get all sorts
of collisions on even medium usage. Full duplex will
more than double your transmission rates. ( The answer
2 and answer 3 talk the same thing with
auto_negotiation). (Thanks Martin Hepworth)

4) Personally I'd leave adv_autoneg_cap on - this
allows the QFE port to tell the switch what
speeds/duplexes it's capable of (as defined by the
other _caps). Otherwise, the rest looks right. (Most
people agreed that 'set auto_negotiation 0).

Thank You All:
Darren Dunham; Debbie, DRusolo at ap.org; Martin
Hepworth; Dinwiddie, Ron; Gaziz Nugmanov; Haywood,
Steven; Jason Premo; Perttunen, Bruce; Steve Mickeler
LAUNCH - Your Yahoo! Music Experience
http://launch.yahoo.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From Sue.Thielen at epeople.com  Thu May 16 14:08:16 2002
From: Sue.Thielen at epeople.com (Sue Thielen)
Date: Thu, 16 May 2002 11:08:16 -0700
Subject: SUMMARY: Tape Drive issues
Message-ID: <3AFA89415D867A4C98808227A0ED53AD1E3BC5@zombie.epeople.com>

I hadn't made any changes to this system at all.. in ages..
but someone must have done something with the SCSI
cable.. I reseated both ends of it, and now the tape drive
is functioning normally..

Thanks to all who replied before I even saw my post !

Thanks.

sue	


>  -----Original Message-----
> From: 	Sue Thielen  
> Sent:	Thursday, May 16, 2002 10:25 AM
> To:	Sun Manager (E-mail)
> Cc:	Sue Thielen
> Subject:	Tape Drive issues
> 
> I'm having an issue with my DLT drive attached to an E 450 running 2.6
> When I try to access a tape, I get the following error messages..
> 
> fortknox# mt -f /dev/rmt/0cn rew
> /dev/rmt/0cn: no tape loaded or drive offline
> 
> fortknox# dd if=/dev/rmt/0c of=/tmp/tape conv=sync 
> dd: /dev/rmt/0c: open: I/O error
> 
> This with a tape in the drive ... 
> 
> In /var/adm/messages I have these messages..
> 
> May 16 10:19:55 fortknox unix: WARNING: /pci at 1f,4000/scsi at 2 (glm1):^M^M
> May 16 10:19:55 fortknox unix:  SCSI bus MESSAGE IN phase parity error
> May 16 10:19:55 fortknox unix: WARNING: ID[SUNWpd.glm.parity_check.6008]
> May 16 10:19:58 fortknox unix: WARNING: /pci at 1f,4000/scsi at 2 (glm1):^M^M
> May 16 10:19:58 fortknox unix:  SCSI bus MESSAGE IN phase parity error
> May 16 10:19:58 fortknox unix: WARNING: ID[SUNWpd.glm.parity_check.6008]
> 
> Any ideas?? Is this a hardware issue??
> 
> Thanks in advance.
> 
> sue
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From Yunics at aol.com  Thu May 16 14:39:31 2002
From: Yunics at aol.com (Yunics at aol.com)
Date: Thu, 16 May 2002 14:39:31 -0400
Subject: SUMMARY: market share of Tape drive technologies
Message-ID: <74A300C1.093E954E.00059E6B@aol.com>

Thanks to the poeple who replied -
Jeff Kennedy, Alex Wilkinson, David Foster and 
FM Taylor.
I could not find exactly what i was looking for but David Foster pointed me towards an excellent comparison of tape technologies at  http://www.spectralogic.com/technology/index.cfm.
I would probably go with AIT over mammoth since it was overwhelmingly preferred by the respondents and there is a clear future roadmap for AIT which mammoth doesnt have yet.

Thanks for ur time.

My original post : 

Admins,
 I know its a little off topic but can anyone point me to a  link/page/document which shows the relatively current market share for different tape drive technologies in the tape backup-restore market. I am specifically looking for "AIT-2" and "Mammoth2" market share to choose between them.

Thanks and will summarize.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From jeff.gilbreath at mirant.com  Thu May 16 16:47:40 2002
From: jeff.gilbreath at mirant.com (Gilbreath, Jeff)
Date: Thu, 16 May 2002 16:47:40 -0400
Subject: SUMMARY: Can't set MTU higher than 1500 on Sun GigE cards
Message-ID: <CF3337B43D42A34EB661E1FA813F68A909A09B@atlexm02>

Thanks to all the people who responded.

Quick Overview:
Private switched but non-routed network with Cisco Catalyst switch and all GigE blades in switch.
Various Sun servers with GigE cards connected to private network.
GigE to GigE, so fragmentation is not a issue.
Testing done through switch as well as direct server to server.

Symptom:
I can't turn the MTU size up on the GigE interface to increase throughput.

Answer:
Official answer finally recieved from Sun is the Sun GigE cards no longer support Jumbo packets.
And can not be set to a MTU higher than 1500 (same as 10BaseT/100BaseT).

Result:
Sun GigE cards max out at about 440Mbits transfer rate even running multiple treaded iperf and ttcp tests.

My response:
Thanks to the people who suggested using SysKonnect and Intraserver cards as a alternative
to the Sun cards. They seem to be higher performance and cheaper too.

Note: The Tolly report thats listed on Sun's web page must be using the previous version.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From linlifeng at yahoo.com  Thu May 16 23:26:25 2002
From: linlifeng at yahoo.com (Lin Feng)
Date: Thu, 16 May 2002 20:26:25 -0700 (PDT)
Subject: SUMMARY: getpwd: permission denied for D1000 disks 
Message-ID: <20020517032625.29197.qmail@web13007.mail.yahoo.com>

Thanks to Frank Smith, Jeff Woolsey and Darren Dunham.

The problem was the origial mount point has more restrict permisions. The solution was to unmount the file systems and change the permision on the mount points and remount the file systems. 

Thanks all very much.

The original quesitons:

> All drives on my D1000 are mounted fine. However I get this error:
>
> $ man cpio
> getcwd: Permission denied
>
> But here is the permission:
>
> $ su
> Password:
># ls -la
> total 22
> drwxr-xr-x   4 dba      osdba        512 May 16 19:31 .
> drwxr-xr-x  47 root     root        1024 May  9 07:38 ..
> drwx------   2 root     root        8192 May  9 08:33 lost+found
> drwxr-x---  14 dba      osdba        512 May  8 20:46 Rel10
>#
>
> (The man command works if I run as root)
>
> When I do a truss on the command, here is what I got:
> ....
> fstat64(3, 0xFFBEF9A0)                          = 0
> getdents64(3, 0x0002A718, 1048)                 = 264
> getdents64(3, 0x0002A718, 1048)                 = 0
> close(3)                                        = 0
> pathconf(".", _PC_PATH_MAX)                     = 1024
> stat64("./", 0xFFBEFA30)                        = 0
> stat64("/", 0xFFBEF998)                         = 0
> open64("./../", O_RDONLY|O_NDELAY)              Err#13 EACCES
> getcwdwrite(2, " g e t c w d", 6)                       = 6
> : write(2, " :  ", 2)                           = 2
> Permission deniedwrite(2, " P e r m i s s i o n   d".., 17)     = 17
>
> write(2, "\n", 1)                               = 1
> llseek(0, 0, SEEK_CUR)                          = 5769
> _exit(1)
LAUNCH - Your Yahoo! Music Experience
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From Ramiro.Santos at commerzbankib.com  Fri May 17 04:59:02 2002
From: Ramiro.Santos at commerzbankib.com (Santos, Ramiro)
Date: Fri, 17 May 2002 10:59:02 +0200
Subject: SUMMARY:  Usable IDE DVD-ROMs with Sun Solaris
Message-ID: <1D1B039860ABD41188C500508BAF2EB2024832A8@xmx9fraib.fra.ib.commerzbank.com>

Hi, sorry for the delay,

one Sun guy told me the only relevant thing is that the drives
musst support data seg. in 512k mode.


Other answers were:

>> Sun uses Hitachi DCD drive.
>> I have used Hitachi GDM-7000 sucessfully


>> I have not heard of anyone trying that, and the Sun System
>> Handbook(which you can find links to from sunsolve.sun.com) does
>> showsthat no Ultra 10:s or 5:s are delivered with DVD. In other words,
>> it is not supported. But "Not supported" != "Doesn't work". (!= not
>> equal)
>>
>> The Sunblade 100 is(was?) delivered with 12x and 16x DVD from Pioneer
>> (Pioneer DVD-115 and DVD-1165M). I wouldn't be suprised if they worked
>> with Ultra 10:s too although I have no heard anyone trying. Maybe
>> booting from them can be a problem but since Solaris can use them un
>> Sunblade 100, it should be able to use it on Ultra 10:s unless there is
>> hardware limitation.
>>
>> But the only way to be sure is to try it.
>>
>> My guess is that either most DVD:s works or none at all.
>>
>> BTW, putting a DVD into a Ultra 10 will leave you with an unsupported
>> configuration even if it works, but I guess you are aware of that.


Many thanks to:
			Krister Bergman
			Hichael Morton
			Heilke Rainer



Yours sincerely,

Ramiro Santos

ZIT P 7.41 Unix Server Implementation

_______________________________________________________

Contact:   Phone:     +49-(0)-69-136-43523
           Fax:       +49-(0)-69-136-47040
           E-Mail:    Ramiro.Santos at CommerzbankIB.com

Address:   Commerzbank AG
           Zentraler Servicebereich IT Production
           C/S Services Investment Banking - ZIT P 7.41
           Mainzer Landstr. 151 - DLZ 1, 06.52.0560
           D-60321 Frankfurt / Main
_______________________________________________________
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From suhas at ctl.creative.com  Fri May 17 05:03:02 2002
From: suhas at ctl.creative.com (Suhas BHIDE)
Date: Fri, 17 May 2002 17:03:02 +0800
Subject: SUMMARY:- largefiles on VxFS
Message-ID: <OF4E63C82D.D44C2274-ON48256BBC.0030FFBA@ctl.creative.com>

Thanks a lot for all who have responded to this.

The answer is to run:

/usr/lib/fs/vxfs/fsadm -o largefiles /filesystem

The only point is that "mount" does not show the "largefiles" but fsadm
will show this.

Thanks a lot to all those listed below:

Amit Mahajan
Wolfgang Kandek
Jason Wood
Mike D. Kail
John Riddoch

Regards...Suhas Bhide
----- Forwarded by Suhas BHIDE/CTLSG/CTL/Creative on 05/17/02 04:55 PM
-----
                                                                                                                                       
                      Suhas BHIDE                                                                                                      
                                               To:       sunmanagers at sunmanagers.org                                                   
                      05/17/02 11:04 AM        cc:                                                                                     
                                               Subject:  largefiles on VxFS                                                            
                                                                                                                                       



Hi All...

I am using VxVM 3.0.4 and VxFS 3.3.3  on our E450 machines. I need to
change
some volumes to support largefiles which are currently not supporting.
I checked the mkfs syntax but could not find anything wherein I can do this
when the filesystem is mounted.
Can anyone help me on this ? Thanks in advance.

Suhas Bhide
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From plamadeleine at lightbridge.com  Fri May 17 09:28:44 2002
From: plamadeleine at lightbridge.com (Paul LaMadeleine)
Date: Fri, 17 May 2002 09:28:44 -0400
Subject: Summary: Backing up a veritas file system questions
In-Reply-To: <5.1.0.14.0.20020516113032.01fa5ec0@postal.lightbridge.com>
Message-ID: <5.1.0.14.0.20020517092059.03005bd0@jupiter.credtech.com>

Well,

         This just goes to show my unfamiliarity with veritas (this is the 
first time I'm deploying it).  When doing a "man -k" for dump,backup,vx I 
never saw any commands to backup the system.  the veritas man pages were 
not in the man path - I had thought that the installation would have taken 
care of that, guess not.  When I went to the veritas books, I looked up 
backup and dump and did not see anything about vxdump (when I looked up 
vxdump it was there though).  Now I know to look things up by keyword with 
a vx in front when looking in the index of those books.

         The anwser from around 50 people so far - use vxfsdump/vxdump.  A 
couple people have said they have used ufsdump successfully as it is a usf 
file system.

         I would like to thank everyone for not blasting me with "read the 
manual" type responses.  Not one negative, degrading response for what 
actually turned out to be a really simple question.

         Thanks,

         Paul


At 11:35 AM 5/16/02 -0400, Paul LaMadeleine wrote:
>Greetings,
>
>         After reading bunches of archives and veritas docs, I can't seem 
> to find my answer.  What I need to do is backup all file systems on a box 
> to a remote tape drive.  We do not have and "official" backup software, 
> just stuff that comes with solaris.  What I was doing before we installed 
> veritas file systems on this server was I was just using ufsdump to dump 
> to a remote tape drive.
>
>         I don't think I can, but I thought I'd ask anyway - can you use 
> ufsdump to dump veritas file systems?
>
>         Is there a better way then tar (or gtar) for backing up these 
> filesystems?
>
>         And we have no money in the budget to purchase any software for 
> doing backups.
>
>         thanks,
>
>         Paul
>_______________________________________________
>sunmanagers mailing list
>sunmanagers at sunmanagers.org
>http://www.sunmanagers.org/mailman/listinfo/sunmanagers
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From ahoesch at smartsoft.de  Fri May 17 11:35:56 2002
From: ahoesch at smartsoft.de (Andreas Hoeschler)
Date: Fri, 17 May 2002 17:35:56 +0200
Subject: Summary: scripting tricks (shell, expect,...)
In-Reply-To: <20020517145719.GH28511@rdlg.net>
Message-ID: <C779555B-69AB-11D6-A947-003065CCA582@smartsoft.de>

Hi all,

thanks a lot for the overwhelming feedback to my question. I got about 
30 answers in 5 minutes. I am still browsing through all the replies. 
Most of you suggested to do something like

#!/bin/bash
ping snap >/dev/null
if [ $? != 0 ]; then
   echo snap is down! Do something about it...
else
   echo snap is already on!
fi

which worked great for me.

Thanks again!

Regards,

    Andreas


=========================================================
Original question:
this is a question for shell script gurus. I need a script called by 
cron that performs a specific action (call another script) if host snap 
is down. I thought about checking this with

bash-2.03# ping snap
no answer from snap

If snap is down the answer "no answer from snap" is returned. But how 
can I use this in a shell or expect script to call my action script? If 
the answer "snap is alive" is returned I want to do nothing. I guess 
this can easily be done but I am unfortunately not such a script expert 
yet.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From john at mtbiker.net  Fri May 17 12:54:18 2002
From: john at mtbiker.net (John Martinez)
Date: Fri, 17 May 2002 09:54:18 -0700
Subject: SUMMARY: NFS performance: Solaris servers and Linux clients
References: <3CE45DC3.4040806@mtbiker.net>
Message-ID: <3CE535BA.9060207@mtbiker.net>

Thanks to:

Broun, Bevan <brounb at adi-limited.com>
Steve Mickeler <steve at neptune.ca>
Amit Mahajan <amahajan at santelnet.com>
Michael Sullivan <mps at discomsys.com>
system administration account <sysadmin at astro.su.se>
Michael Hase <michael at six.de>
Thomas Anders <anders at hmi.de>
Francisco Puente <francisco.puente at cea.ericsson.se>
Christophe Dupre <duprec at scorec.rpi.edu>
Wesley Suess <wes.suess at bigidea.com>
Justin Shaffer <Justin.Shaffer at mlb.com>
Fabrice Guerini <fabrice at bluemartini.com>
Tim Chipman <chipman at ecopiabio.com>

Original question:
 >
> My set up:
> 
> NFS Server
>   Sun E4500 8x400MHz processors, 8GB RAM
>   Solaris 8
>   VxVM 3.2
>   VxFS 3.4
>   Gigabit ethernet
>   SAN storage
> 
> Accessing from Solaris clients is fast for both read and writes.
> 
> Linux clients have fast reads, but slow writes. All clients with 
> problems are standard RedHat 6.2 and RedHat 7.2 boxes.
> 
> Anything I can do on my Solaris servers to try and troubleshoot? 
> Everything inside me tells me that Linux is the culprit, but I'm not 
> 100% sure.
> 
> thanks,
> -john

I received answers of various kinds. All agreed that the problems were 
with the Linux clients, not the Solaris server. I have not tried any of 
the suggestions yet, but will try that over the next few days. I wanted 
to summarize the responses before the weekend.

Various answers include:

   o  Patching the Linux kernels to the latest 2.2 and 2.4
      variations, which support NFSv3 and mounting NFS over
      TCP

   o  Increasing the write and read block sizes to 8k on the
      Linux NFS clients with the mount options:

         rsize=8192,wsize=8192

         -or-

         rsize=32768,wsize=32768

      Some suggested going up to 32k, which is standard in
      Solaris NFS mounts

   o  Some suggested forcing TCP mounting after patching the
      Linux kernel to support it with the following option to
      mount:

         tcp

   o  Some suggested forcing NFSv3 mounting with the following
      mount option on the client:

         nfsvers=3

   o  Another suggestion was to patch the Linux kernel with
      Trond Myklebust's NFSv3 patches at

         <http://www.fys.uio.no/~trondmy/>

   o  Sample mount commands include the following (sorry for
      wrap-around):

         solaris.nfs.server:/nfs/share/path   /local/mount/point  nfs 
rsize=8192,wsize=8192,timeo=14,intr,tcp 0 0

         server:/export /mountpoint nfs 
bg,soft,tcp,rsize=32768,wsize=32768,nfsvers=3 0 0



Thanks to all who responded.

-john
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From fabrice at bluemartini.com  Fri May 17 15:04:32 2002
From: fabrice at bluemartini.com (Fabrice Guerini)
Date: Fri, 17 May 2002 12:04:32 -0700
Subject: SUMMARY: Determine Usable Disks at JumpStart
In-Reply-To: <3CE41FCE.3080709@emc.com>
References: <5.1.0.14.0.20020516114012.02217db8@mail.bluemartini.com>
Message-ID: <5.1.0.14.0.20020517115616.02617fe8@mail.bluemartini.com>

Thanks to: Bryan Moore, Kevin Graham, Darren Dunham, Glenn Cantello, David 
Foster, Tim Fritz, Bryan Pepin.

Nobody really provided a way to tell writable disk devices from read-only 
devices, but all the suggestions pointed me to a simple solution that 
solved my current problem, which was to determine the boot disk at 
JumpStart time.

The answer: let JumpStart determine it itself. The boot disk can then be 
referred to in the rest of the profile as "rootdisk". This resulted in the 
following profiles:

For initial installation,

    install_type   initial_install
    system_type    server
    boot_device    any             update
    partitioning   explicit
    filesys        rootdisk.s0     free    /
    filesys        rootdisk.s1     2048    swap
    filesys        rootdisk.s3     128
    cluster        SUNWCXall

For upgrade,

    install_type   upgrade
    cluster        SUNWCXall

My original question:

>>I am trying to programmatically generate a profile for JumpStart 
>>installations. The issue, here, is that I cannot predict the disk that 
>>will be used for O.S. installation.
>>
>>In 99% of cases, I will be able to use "c0t0d0s0", but there are fringe 
>>cases where the actual disk where I put my root partition is on a 
>>different device name (case in point, my SunFire 280R is installed on 
>>"c1t0d0s0", while the device on controller 0 is the CD-ROM drive).
>>
>>I know I can determine the list of mountable devices using "iostat -xpn", 
>>but how can I detect whether the device is a writable disk, suitable for 
>>Solaris installation, or if it isn't (CD-ROM drive or other read-only 
>>device)? The "format" command seems to be able to do that.
>>
>>Thanks in advance.



+===========================================================+
| Fabrice "Script It!" Guerini  Blue Martini Software, Inc. |
| Senior Operations Engineer    2600 Campus Drive           |
| Tel: (650) 356-7576           San Mateo, CA 94403-2522    |
| Fax: (650) 356-4001           www.bluemartini.com         |
+===========================================================+
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From hford at globix.net  Fri May 17 16:20:17 2002
From: hford at globix.net (Harry Ford)
Date: Fri, 17 May 2002 16:20:17 -0400 (EDT)
Subject: SUMMARY: SunVTS: vtsk: svc_reg: Bad file number
In-Reply-To: <Pine.LNX.4.33.0205151445260.7369-100000@hford.globix.net>
Message-ID: <Pine.LNX.4.33.0205171619390.25291-100000@hford.globix.net>

The solution is:

  run rpcbind.


thank you for your help, everyone.

-Harry



On Wed, 15 May 2002, Harry Ford wrote:

>
>
> Hi All,
>
> Trying to run Sun VTS on a netra (Sol 8, developer install, 04/01,
> VTS-4.3) and I'm getting the following error:
>
> root at web1:/opt/SUNWvts/bin# ./sunvts -t
> vtsk: svc_reg: Bad file number
> sunvts: Can't start vtsk properly. Stop.
>
> any ideas?  will summarize.
>
> -Harry
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From alan at motorola.com  Fri May 17 17:23:01 2002
From: alan at motorola.com (McIntosh Alan-R54071)
Date: Fri, 17 May 2002 14:23:01 -0700
Subject: SUMMARY: Configure system as an SSP for an E10000 
Message-ID: <D3FB80858D62D511ACD500D0B73E9D8502367386@az33exm03.corp.mot.com>

That was easy!!  Thanks to Alex Shepard for his quick
response of:

	Make sure you're not installing the SSP package from the 
	"Computer Systems Supplement CD" during your install.  

	You don't mention which type of install you're doing, but 
	I've seen this on web installs, where you have to go in 
	and de-select SSP from the list of products.

I normally just do a "Default Install" on any system that
is not short on disk space, the two with this error it was
necessary to perform a "Custom Install".  Why the SSP software
is not part of the "Default Install", and must be 'de-selected'
from the "Custom Install", is beyond me.

-Alan


-----Original Message-----
From: McIntosh Alan-R54071 [mailto:alan at motorola.com]
Sent: Friday, May 17, 2002 1:53 PM
To: Sunmanagers (E-mail)
Subject: Configure system as an SSP for an E10000 


Sunners,

I have received this message on the boot sequence of a 
system that was just loaded with Solaris 8 from a fresh 
install: 

Do you wish to configure this system as an SSP for an E10000? (y/n) 

This is the second time that this has occurred on a fresh 
install, it has never appeared when doing an upgrade. It is
strange that it has only occurred twice out of the many times
I have done a clean install.  The system will just hang until 
a response is given. 

How can I get rid of this message? 

-Alan
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From ak94601 at yahoo.com  Fri May 17 17:57:47 2002
From: ak94601 at yahoo.com (Ath Kar)
Date: Fri, 17 May 2002 14:57:47 -0700 (PDT)
Subject: Summary: SunFire 880 SWAP
Message-ID: <20020517215747.44808.qmail@web14801.mail.yahoo.com>

Hi 

Thanks once again. I was confused with df -k output.
As this is my first time to work on Solaris 8, I saw 
swap ..mounted on /var/run first time. And secondly, I
overlooked swap is displayed in "blocks" not in
Kbytes.

Thanks to Michael, Darren, Tim, John, Scott, Jed and
Simon, Bertrand, Larye, Darren, Kachan, Todd who
solved my firs query... My summary could not be
distributed (why ? I donot know the reason ..Moderator
approval ?) Hence I am giving Summary again..

We can change primary swap by adding temporary swap
and then by changing partition for original swap and
activate it again as follows:

1. makefile 8g /export/home/swapnew (make new swap
   file for temporary perpose)
2. swap -a /export/home/swapnew     (to add swap)
3. swap -l                          (to check swap)
4. swap -d /dev/dsk/<swap partition> (disable swap)
5. format (make a swap partition as desire)
6. swap -a /dev/dsk/<swap partition>
7. swap -d /export/home/swapnew (added for tmp
perpose)
8. Also change the entry for swap in /etc/vfstab, if
you have defined different partition)

Thats it

Ath
8. 
LAUNCH - Your Yahoo! Music Experience
http://launch.yahoo.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From sreenureddy24 at yahoo.com  Sat May 18 01:04:50 2002
From: sreenureddy24 at yahoo.com (sreenu reddy)
Date: Fri, 17 May 2002 22:04:50 -0700 (PDT)
Subject: "SUMMARY: Need of a script"
Message-ID: <20020518050450.32720.qmail@web20408.mail.yahoo.com>

Hello managers,

              I am sorry a little bit late in posting
the summary ,because of lot of work in the job. I got
more than 30 replies , every script is working.Thanks
to every one.  This is the first script I got from
steinaxt at bajazzo.at , so I followed it.Once again
thanks to everyone.
--- cut here ---
#!/bin/sh
DIR1=/var/spool/mqueue.tmp
DIR2=/var/spool/mqueue

for qfile in ${DIR1}/q*;do
        file=`basename ${qfile} | sed 's/^q//'`
        mv ${DIR1}/[qxs]${file} ${DIR2}/
        sleep 10
done
--- cut here ---
LAUNCH - Your Yahoo! Music Experience
http://launch.yahoo.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From juniorunix at hotmail.com  Sat May 18 04:55:13 2002
From: juniorunix at hotmail.com (mr ayie)
Date: Sat, 18 May 2002 16:55:13 +0800
Subject: SUMMARY:how to change back my device links w/out reboot
Message-ID: <F69sicvT2JJnOolq1G00001e62b@hotmail.com>

hi thanks to

bertrand hutin,john england,neill mark & homan charles for responding.

although some of the still insist me to use devfsadm or drvconfig that i 
have tried before only bertrand hutin that suggest to me something so simple 
that i haven't though before simply by mv /dev/rmt/1* to 0* & /dev/rmt/0* to 
1* ..

the netbackup is running fine now and so the external tape drive. thanks 
again to them expecially bertrand.







_________________________________________________________________
Join the worlds largest e-mail service with MSN Hotmail. 
http://www.hotmail.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From RABENOJA at prodigy.net  Sat May 18 11:09:45 2002
From: RABENOJA at prodigy.net (RABENOJA at prodigy.net)
Date: Sat, 18 May 2002 10:09:45 -0500
Subject: Summary: Remote Access and Headless
Message-ID: <AA-B5F6960A56219FE1070B44EE1FED1249-ZZ@homebase1.prodigy.net>

I want to thank everyone who responded.  Everyone came 
up with # eeprom auto-boot?=true which in our case 
was already set , but it was Jaime Dela Rosa who came 
up with the little bit extra eeprom setting on ttya 
that fixed our problem.

Here is his response:

most likely your eeprom setting has the auto-boot?
=false. change it to true.

i.e. inside Solaris:

# eeprom auto-boot?=true

and/or check these variables as well:
ttya-rts-dtr-off=false
ttya-ignore-cd=true


i hope this helps


Regards,


---------------
Jaime Dela Rosa
Siemens Business Services - Australia
e-mail: jaime.delarosa at siemens.com.au 

> -----Original Message-----
> From: RABENOJA at prodigy.net 
[mailto:RABENOJA at prodigy.net]
> Sent: Saturday, 11 May 2002 04:05
> To: sunmanagers at sunmanagers.org
> Cc: ra7839 at pbis.com
> Subject: Remote access to headless systems
> 
> 
> Dear Sun Gurus,
> 
> We have been administering many headless sparcs out 
in 
> the field. We have ttya connected to a terminal 
> server (which is nothing more than a cisco router) 
so 
> that we can monitor and control these systems from a 
> remote location. We have recently noticed that the 
> newer systems (new for us) with an AXi motherboards 
> seem to have problems when we reboot. When we telnet 
> into the terminal server to access the console, and 
> reboot the system, we seem to lose access to the 
> console as it reboots. We cannot see any data 
> streaming across nor can we enter any keystrokes to 
> enter any commands. In order to recover, I have to 
> dispatch a Site Manager and have them plug in a 
> keyboard and a monitor to bring the system back up. 
I 
> have been told that these systems are at the ok 
prompt 
> and then I have to get the sight manager to type 
boot 
> and remove the keyboard during ram check.
> 
> We are running some proprietary software on these 
> systems, but as I have stated earlier the older 
> systems with sparc motherboards (ultrasparc  Ultra 
> 1 s) are working fine and have no problems after 
> rebooting.
> 
> If anyone can give me some tips as where to start 
> troubleshooting I would appreciate it.
> 
> Thank you,
> Robert Abenoja
> _______________________________________________
> sunmanagers mailing list
> sunmanagers at sunmanagers.org
> 
http://www.sunmanagers.org/mailman/listinfo/sunmanagers
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From markhj01 at bahamut.thelair.dynamic-site.net  Sat May 18 11:31:32 2002
From: markhj01 at bahamut.thelair.dynamic-site.net (Dragon King)
Date: Sat, 18 May 2002 11:31:32 -0400 (EDT)
Subject: SUMMARY: Linux <-> Solaris NFSv3 not working
In-Reply-To: <C1A9CBC91833D611B6740002B31015D07901@SUNEXCH3>
Message-ID: <Pine.GSO.4.10.10205181124550.7640-100000@bahamut>

Much thanks for the help from:
Rick Kelly
Francisco Puente
Michael DeSimone
Josh Glover
Michael Sullivan
David Foster
Rik Schneider

Unanimous response was look at the kernel to make sure support was
compiled in, some said kernel 2.4.. Rik Schneider also pointed out
nfs.sourceforge.net for NFSv3 patches for the 2.2.19 kernel. That set me
on the correct path.


From psharmaus at yahoo.com  Sun May 19 11:24:34 2002
From: psharmaus at yahoo.com (P Sharma)
Date: Sun, 19 May 2002 08:24:34 -0700 (PDT)
Subject: SUMMARY:Max. Hard Drive Compatible for ultra10
Message-ID: <20020519152434.12374.qmail@web14908.mail.yahoo.com>

Hi,

I thank Paul for his response: 

I have a Maxtor 27gb and an IBM GXP 60 - 60gb disk fitted 
in an Ultra 5 running Solaris 8.
The ultra 5 and ultra 10 are same m/b in a different case.
I dont think you can use large disks like these with earlier 
Solaris versions.

I also thank Serge Borkov for his response about format, cyl/heads.

Thanks a lot

Sharma
LAUNCH - Your Yahoo! Music Experience
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From erwin.mendoza at cronusgroupinc.com  Sun May 19 22:29:56 2002
From: erwin.mendoza at cronusgroupinc.com (Erwin B. Mendoza)
Date: Mon, 20 May 2002 10:29:56 +0800
Subject: SUMMARY: DNS Procedure
In-Reply-To: <193130-22002511331746562@cronusgroupinc.com>
Message-ID: <NGBBKAGLIMLIEHOMEBNJAEHFCEAA.erwin.mendoza@cronusgroupinc.com>

Hello,

I would like to apologize for the late response in posting the SUMMARY of this
topic. I was not in the office for couple of weeks.
Anyway, I want to thank everyone who responded on this topic. It came to me a
very helpful for every suggestions and tips. Some gave me a link or a suggestion
to download a bind in sunfreeware.

Here are some links:
http://www.ebsinc.com/solaris/dns.html
http://www.nominum.com/resources/documentation/Bv9ARM.pdf
http://www.solarisguide.com/news_story.php3?ltsn=2001-01-20-016-02-QG

Thanks and more power.


-----Original Message-----
From: sunmanagers-admin at sunmanagers.org
[mailto:sunmanagers-admin at sunmanagers.org]On Behalf Of Erwin B. Mendoza
Sent: Monday, May 13, 2002 11:18 AM
To: sunmanagers at sunmanagers.org
Subject: DNS Procedure


Hello Everyone,

I'm here asking some questions.... :-)

I would like to ask if anyone could give some links or files on procedure
in setting up a DNS.

It's like having a new solaris box that I will make it a web server.

Any help would highly appreciated....

Thanks and more power....


Regards,
ERWIN
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From khalil.hammid.b at us.agfa.com  Mon May 20 09:03:38 2002
From: khalil.hammid.b at us.agfa.com (khalil.hammid.b at us.agfa.com)
Date: Mon, 20 May 2002 09:03:38 -0400
Subject: SUMMARY: 18GB Drives in D1000 and E450
Message-ID: <85256BBF.0047B38F.00@AGWJ0X.AGFA.BAYER.COM>

Hello Everyone,

First, I would like to apologize for the delay in submitting my summary.  Here
was my original query:

> Hi All,
>
> Are there any compatibility issues/problems if I replace a 18GB 10K drive in a
> D1000 with a 18GB 10K drive taken from an E450 server?  I looked up the part
> numbers from SunSolve and got the following description:
>
> D1000 Drive    540-4178-01     DRV 18GB 10K 1-in SCSI W/SPUD&PLT        -
>
> E450 Drive     540-4177-01     DRV ASSY 18GB 10K 1-in SCSI W/SPUD        -
>
> I have extra drives in the E450 and would like to use one of those to replace
a
> failed drive in the D1000.  Sun Tech Support told me unsupported hardware
> (meaning the drive taken from the E450) could cause problems in the D1000. He
> did not say exactly what problems  this can cause but he suggested we order a
> drive with the correct part number.
>
> Thanks,
>
> Khalil

The fact is, although the D1000 and E450 drives may look the same, they are
constructed differently.  The D1000 uses differential (HVD) scsi drives. While,
the E450 uses LVD/SE scsi drives internally. The two types are not compatible
with one another.  It may work, but eventually you may run into some problems.
And of course, Sun Service will not provide service for unsupported hardware.

Thanks to Paul Grieggs, Peter Stokes, Jeff Thomas, Tony Walsh, Stan Pietkiewicz,
Matt D. Harris, Matthew Stier, David Fung, Chris Price, Krister Bergman and
Mohammad Saqib for your quick replies.

Regards,

Khalil Hammid
Product Support Specialist
Technical Center
Agfa Corporation
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From thesunlover2002 at yahoo.com  Mon May 20 09:53:40 2002
From: thesunlover2002 at yahoo.com (John Lee)
Date: Mon, 20 May 2002 06:53:40 -0700 (PDT)
Subject: SUMMARY: Will "ndd -set adv_autoneg_cap 1" ignore the other ndd
  commands  running before it?
In-Reply-To: <20020516182830.47515.qmail@web14902.mail.yahoo.com>
Message-ID: <20020520135340.39950.qmail@web14907.mail.yahoo.com>

Thanks: Anthony Miller
(anthony.miller at vf.vodafone.co.uk), "Darren Dunham"
<ddunham at taos.com>, "Homan, Charles (NE)"
<Charles.Homan at GDC4S.Com>, McCaffity, Ray
(Contractor)"<McCaffityR at epg.lewis.army.mil>,
Simon-Bernard Drolet" <sxi at sympatico.ca>

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Question:

From nick.hindley at lbhf.gov.uk  Mon May 20 11:28:27 2002
From: nick.hindley at lbhf.gov.uk (Hindley Nick)
Date: Mon, 20 May 2002 16:28:27 +0100
Subject: SUMMARY: T3 Disk array
Message-ID: <02May20.162919bst.119046@lbhfinet.lbhf.gov.uk>

All documentation for this lives at


http://sunsolve.sun.com/handbook_pub/Systems/T3/docs.html?wrapper=false

has all of them, including the disk administrators guide in .pdf format.



The information contained in this e-mail is intended for the recipient or
entity to whom it is addressed. It may contain confidential information that
is exempt from disclosure by law and if you are not the intended recipient,
you must not copy, distribute or take any act in reliance on it. If you have
received this e-mail in error, please notify the sender immediately and
delete from your system. 
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From sunman at ureach.com  Mon May 20 11:41:46 2002
From: sunman at ureach.com (Vahid Moghaddasi)
Date: Mon, 20 May 2002 11:41:46 -0400
Subject: SUMMARY: Slow performance on EMC
Message-ID: <200205201541.LAA11820@www21.ureach.com>

Hi everyone,

I received only one reply from John Sellin which was comforting in regards to the 
following question I had posted:

How easy is it to upgrade EMC Powerpath 1.3 to 1.5 and VXFS from 3.3.1 to 3.4 
p2?
John's response:
Both were/are very simple upgrades.  Since we had to have the environment down, 
we went ahead and upgraded vxva from 3.0.4 to v3.1.  The vx stuff is just pkgrm's 
and pkgadd's.

I must add that I upgraded both this past weekend and it was a walk in the park!
Reboots took more time than the actual upgrade. All were done in 3hours.
Thank you,
Vahid.

________________________________________________
Get your own "800" number
Voicemail, fax, email, and a lot more
http://www.ureach.com/reg/tag
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From FiamingF at strsoh.org  Mon May 20 13:47:14 2002
From: FiamingF at strsoh.org (Fiamingo, Frank)
Date: Mon, 20 May 2002 13:47:14 -0400
Subject: SUMMARY - Serial Console Connectivity
Message-ID: <86BAB17B6CE9D511BFDA00508BAC9EB3BC8CF9@marius.strsoh.org>

Thanks to all who responded.  The consensus of responder who use
Cyclades equipment is that it works very well, with the possible exception
of connectivity to Sun V880.  One user reported that reboots of the V880,
but
not other servers, could cause the terminal server to crash.  The Cyclades
was 
recommended for its:
	ssh connectivity
	ability to have multiple connections to the same port (1 active,
others passive)
	controlled access to ports (with user accounts), and
	ability to send a "break" only when desired (programmable).

Other positive recommendations were from users with:
NetReach http://www.wti.com/netreach.htm
Cisco 2511 and 36[x]0
Aurora Technology's Control Tower  www.auroratech.com 
HP Web Console
Perle (www.perle.com)
Lightwave Inc product Console Server 3200 www.lightwavecom.com

	Thanks,
	Frank

> -----Original Message-----
> From: Fiamingo, Frank [mailto:FiamingF at strsoh.org]
> Sent: Monday, May 13, 2002 10:07 AM
> To: 'sunmanagers at sunmanagers.org'
> Subject: Serial Console Connectivity
> 
> 
> We're looking to connect about 3 dozen Sun server console
> ports to some type of network accessible controller.  Two products
> we've heard of are Digiboard and Cyclades, but we don't know 
> very much about either product.  
> 
> Does anyone have any Pros/Cons to report about these or similar
> products?  What do you use and do you like it?
> 
> 	Thanks,
> 	Frank
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From epa9206 at epa.state.il.us  Mon May 20 13:54:20 2002
From: epa9206 at epa.state.il.us (Donald Bricker)
Date: Mon, 20 May 2002 12:54:20 -0500
Subject: Summary:Migrating from Sparc II to Sparc III technology
Message-ID: <sce8f206.008@admisd3.epa.state.il.us>

Thank you to the following for responding:
Jim.R.Jones
F.M. Taylor
Hichael Morton
Hendrik Visage
Ryan Larson
Tim Chipman
John D Groenveld
Al Hopper
Michael Maciolek
christopher.hubbell

Lots of things to consider. The overall consensus was to use the current
systems (SparcII) as long as possible. Meaning as long as they perform their
function and are not at thier end of life cycle.

Thanks again


Don Bricker
Database Administrator / System Administrator
Illinois Environmental Protection Agency
1021 North Grand Avenue East
Mail Code #32
Springfield, IL 62794-9276
Donald.Bricker at epa.state.il.us
(217) 558-2290
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From topher at findtopher.com  Mon May 20 14:17:27 2002
From: topher at findtopher.com (topher)
Date: 20 May 2002 14:17:27 -0400
Subject: SUMMARY: error "fru: Device busy"
In-Reply-To: <m2vg9ihh3r.fsf@findtopher.com>
References: <m2vg9ihh3r.fsf@findtopher.com>
Message-ID: <m27klyhefc.fsf@findtopher.com>

Many thanks to:
Joshua Fielden
Justin Stringfellow
Alan Fox

The solution that I took was to kill the sf880drd daemon (isn't that redundant?) using /etc/init.d/sf880dr stop and then moving S77sf880dr to s77sf880dr.  It would appear that sf880drd is the 'magic' behind the little V880 push-buttons and LEDs, so my thought would be to start it back up before trying any PCI hot-plugging/swapping actions...

This is apparently a known issue with V880s right now (not at all related to the T3) and has no known solution (so I most likely would have gotten it from sunsolve if I could get there)

Joshua had a problem that turned out to be the CPU/Mem connection quality - and they replaced the backplane and CPU board...

Thanks all!

toph


topher <topher at findtopher.com> writes:

> Greetings all -
> 
> I've got a V880 running Solaris 8 (with the 4/02 Patch Cluster installed) connected to a T3 running firmware revision 2.00.01 (T3 is configured RAID5 w/o hot spare)
> 
> Of course I can't get to sunsolve.sun.com right now to get a later firmware revision for the T3 (keep getting Connection timed out errors) - and all other sources of information have failed me (including the FAQs and the archives).  I'm not completely certain that it's related to the T3, it could be the internal disk - but I can't be certain at this point....
> 
> The problem is a constantly scrolling console error of 'fru: Device busy' - there is no further information in /var/adm/messages or /var/log/syslog - but the error is scrolling by so fast, it's pretty much making the serial console unusable.
> 
> there is an occasional error of "WARNING: bogus error: 0" - but I'm pretty sure that's Java related...  Just don't want to leave out any pertinent information.
> 
> Your thoughts and consideration are appreciated.
> 
> topher
> _______________________________________________
> sunmanagers mailing list
> sunmanagers at sunmanagers.org
> http://www.sunmanagers.org/mailman/listinfo/sunmanagers
> 
> 
> 

-- 
Blessed are the Geeks, for they shall internet the earth.

PGP Key fingerprint = 078E F3BA CA0E 0790 9992 68F8 31E7 B967 5AB3 AC11
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From rickv at mwh.com  Mon May 20 16:51:45 2002
From: rickv at mwh.com (Rick von Richter)
Date: Mon, 20 May 2002 13:51:45 -0700
Subject: [SUMMARY] 'set' vs 'env'
Message-ID: <3CE961E1.B2ADDD17@mwh.com>

Thank you all for the responses.  BTW, I did RTFM and the pages said
nothing explicitly between the two.

Original Question
-----------------
Kinda academic question but what is the difference between the 'set' and
'env' commands.  Typing in either command (without parameters) spits out a
bunch of environment variables.  Is there a difference?

Answer
------
"set" gives both environment and shell variables.  "env" only gives
environment variables, which are passed along to child processes.

-- 
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  Rick von Richter Production Support Manager       Voice: 858-831-2222
  rickv at mwh.com    Maintenance Warehouse/Home Depot   Fax: 858-831-2221
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  The box says: Win98, WinNT or BETTER. That's why I installed Linux.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

[demime 0.99c.7 removed an attachment of type text/x-vcard which had a name of rickv.vcf]
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From jrodriguez at intellinet-tech.com  Mon May 20 17:00:50 2002
From: jrodriguez at intellinet-tech.com (Jeremy Rodriguez)
Date: Mon, 20 May 2002 17:00:50 -0400
Subject: SUMMARY: adding disks
Message-ID: <HMEIJMEKLHGNJPCDCEBOAEMJDAAA.jrodriguez@intellinet-tech.com>

Thank you to Francisco Puente, Graham Wood, Tim Chipman

Question:
I have a raid box A1000, running RAID V, Solaris 7, in 32-bit mode. My
/export/home is 98% full. I will be adding 4-18g disks, which will give
another 32g. I want them to be part of the /export/home partition. How do I
do this? format, newfs, vfstab? Step-by-steps would be great. This RAID box
is attached to a Netra 1125 with no disks. The Netra uses the A1000 as it's
file system.

Answer:
add the new disk to you mdX device like this:

#metattach d0 /dev/dsk/c1t0d0s0 (your new slice)
#growfs -M /export/home /dev/md/rdsk/d0 (And you get your new extra space!!)

or

-full backup of filesystem on the a1000 raid5 LUN
-delete this LUN via GUI RaidManager interface
-create a new LUN containing all the old disks, plus the 4 new 18 gig disks.
(Presumably your existing drives are also 18 gigs/)
-newfs this new, large LUN, mount it to the same place on your filesystem,
ie
/export/home, and then do a full restore from backup. You now have more free
space on the slice :-)
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From khalil.hammid.b at us.agfa.com  Tue May 21 09:32:49 2002
From: khalil.hammid.b at us.agfa.com (khalil.hammid.b at us.agfa.com)
Date: Tue, 21 May 2002 09:32:49 -0400
Subject: Summary (Corrected): 18GB Drives in D1000 and E450 
Message-ID: <85256BC0.004A6229.00@AGWJ0X.AGFA.BAYER.COM>

Hello Everyone,

One of our colleagues who happens to work for Sun Microsystems responded to my
previous summary with the following:

"The Part Numbers you have quoted in your original email are identical 18GB 10K
RPM Single Ended drives with the single exception of an aluminium back plate
attached to the 4278 PN to act as a heat sink. There is no such thing as an HVD
disk drive, therefore the D1000 and the E450 use the same type of Single Ended
drive internally. The only differential item in either the D1000 or E450 is the
controller card of the D1000 or the possible inclusion of a Diff SCSI controller
card in an E450 to attach a D1000 (or an A1000).

Please correct this inaccurate detail at your soonest opportunity."

Here is my original inquiry:

Hi All,

Are there any compatibility issues/problems if I replace a 18GB 10K drive in a
D1000 with a 18GB 10K drive taken from an E450 server?  I looked up the part
numbers from SunSolve and got the following description:

D1000 Drive    540-4178-01     DRV 18GB 10K 1-in SCSI W/SPUD&PLT        -

E450 Drive     540-4177-01     DRV ASSY 18GB 10K 1-in SCSI W/SPUD        -

I have extra drives in the E450 and would like to use one of those to replace a
failed drive in the D1000.  Sun Tech Support told me unsupported hardware
(meaning the drive taken from the E450) could cause problems in the D1000. He
did not say exactly what problems  this can cause but he suggested we order a
drive with the correct part number.

Here is my previous summary:

The fact is, although the D1000 and E450 drives may look the same, they are
constructed differently.  The D1000 uses differential (HVD) scsi drives. While,
the E450 uses LVD/SE scsi drives internally. The two types are not compatible
with one another.  It may work, but eventually you may run into some problems.
And of course, Sun Service will not provide service for unsupported hardware.

Sincerely,

Khalil
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From rrugman at sonusnet.com  Tue May 21 10:55:58 2002
From: rrugman at sonusnet.com (Rugman, Rob)
Date: Tue, 21 May 2002 10:55:58 -0400
Subject: SUMMARY: IP flaky following a reboot.
Message-ID: <4CF48720B6E9D4118D040060CF2074E9034F7FBA@sonusdc3.sonusnet.com>

Thanks go to the following:

Randy Romero, Larye Parkins and Phil Rainey who pointed out that the server
will attempt a reverse lookup on the IP of connecting Telnet clients hence
the huge delay in logging on.

Scott Buecker who suggested a look at the output of netstat -I ge0 which
showed up a not insignificant 1% inbound packet errors, this widened my
search towards more general network errors and this was further backed up by
complaints of abysmal performance of file operations to and from NFS mounts
on the server.  

The final solution to the problem was:

1. Force both the network switch and the ge device to use 1000Mbps Full
Duplex 

2. Tweak the TCP settings using ndd (and a few extra lines in
/etc/init.d/inetinit to preserve the changes on restart) to more suitable
defaults for a gigabit card (thanks to the NetApp site for that one)

3. Apply the latest available patches for the ge device, tcp, and NFS


-----Original Message-----
From: Rugman, Rob 
Sent: 17 May 2002 14:53
To: sunmanagers at sunmanagers.org
Subject: IP flaky following a reboot.



I shutdown/restarted an E4500 this morning.  Although initially everything
looked fine when it came back up, DNS resolution and Telnet (in and out)
both seem very flaky.  DNS simply refuses to resolve addresses while Telnet
seems to work for some users and not for others.  Other login methods e.g
RLOGIN are fine.
 
Any thoughts gratefully received.  I'm busily adding hosts entries as a DNS
workaround :)
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From egonle at netscape.net  Tue May 21 12:34:50 2002
From: egonle at netscape.net (egonle)
Date: Tue, 21 May 2002 12:34:50 -0400
Subject: SUMMARY: E250 cpu module upgrade
Message-ID: <04ABEDDF.0E460BDD.00060FD2@netscape.net>

Hi,

thanks to anybody who responded to my e-mail.

400MHz CPU module need to have a different setting of J3001. 300MHz modules have this jumper set on 1-2, 400MHz modules on 2-3. Location of J3001: If you are looking from front to back.  It is located in front of the memory banks near bank A (left most bank) (thanks to lbmoore(at)scrippsops.com.
If you are going to install two modules on one machine (which I did not intend to do) all modules have to be of them same speed (MHz and cache-> part number).

Regards,
Egonle


__________________________________________________________________
Your favorite stores, helpful shopping tools and great gift ideas. Experience the convenience of buying online with Shop at Netscape! http://shopnow.netscape.com/

Get your own FREE, personal Netscape Mail account today at http://webmail.netscape.com/
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From scot.needy at corp.infi.net  Wed May 15 14:16:41 2002
From: scot.needy at corp.infi.net (Scot Needy)
Date: 15 May 2002 14:16:41 -0400
Subject: SUMMARY: SUMMARY: Loading perl modules bassed on Solaris uname
In-Reply-To: <1021470821.1263.23.camel@sywn001.engrs.infi.net>
References: <1021406764.1227.67.camel@sywn001.engrs.infi.net> 
  <1021470821.1263.23.camel@sywn001.engrs.infi.net>
Message-ID: <1021486629.1261.65.camel@sywn001.engrs.infi.net>

And the winner is 'BEGIN' thanks for the feedback. 
This code will add a search path based on OS Rev. 
BEGIN {
   use POSIX qw(uname);
   my ($uname_s, $uname_r)  = (POSIX::uname())[0,2];
   unshift(@INC, "/var/opt/modules/$uname_s/$uname_r" );
}

If you wish to follow the rest of this problem I have been asked to 
take it off this list so I am now posting to beginners at perl.org
	Thread : DynaLoader and lib problems. 

The above code works unless your module uses DynaLoader to load a
sharred obj. For some reason it isn't searching 
$dir/$archname/auto as documented in `perldoc lib`

Module I am using is Solaris::Kstat

Scot 

 

On Wed, 2002-05-15 at 09:53, scotrn wrote:
> Thanks to those who responded. 
> 
> There are a few discussons on the net about this. Apparently the use 
> statements are processed before ANY other perl code so... 
> 
> use lib $path is equal to use lib "NULL". 
> You cannot use a variable yet because perl has not parsed that bit of
> code that defines $path. 
> 
> The only way I am able to push a dynamic directory onto the INC array is
> by wrapping perl in a shell script setting LD_LIBRARY_PATH or 
> by adding a bit of ugly code. Still can't seem to get DynaLoader to look
> in a different directory so I created a sun4-solaris/auto directory under 
> each 5.6,5.7 amd 5.8 directory to hold my .so files  
> 
> use strict;
> use POSIX qw(uname);
> my ($uname_s, $uname_r)  = (POSIX::uname())[0,2];
> 
> if (($uname_s eq "SunOS") && ($uname_r eq 5.6)) {
>       use lib qw(/var/opt/modules/SunOS/5.6);
>    } elsif (($uname_s eq "SunOS") && ($uname_r eq 5.7)) {
>       use lib qw(/var/opt/modules/SunOS/5.7);
>    } elsif (($uname_s eq "SunOS") && ($uname_r eq 5.8)) {
>       use lib qw(/var/opt/modules/SunOS/5.8);
>    } else {
>       die("$0 is only supported on Solaris 2.6, 2.7 & 2.8 \n");
>    }
> 
> Scot 
> 
> On Tue, 2002-05-14 at 16:06, scotrn wrote:
> > Hi; 
> > 
> >  I was wondering if any of you have tried to dynamicly load 
> > a module from a directory based on the rev of OS? 
> > 
> > In other words....
> > my ($uname_s, $uname_r) = split('\s',`uname -sr`);
> > 
> > use lib "/share/local/modules/$uname_s"."/"."$uname_r";
> > 
> > This seems to work but modules which use DynaLoader to load shared
> > objects it's falling back to the original @INC and not looking in 
> > my lib path. 
> > 
> > Trying to support an NFS share which has multiple versions of compuled 
> > modules bassed on OS Rev. 
> > 
> > Thanks 
> > Scot 
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From RMcEwan at netopia.com  Tue May 21 17:51:27 2002
From: RMcEwan at netopia.com (McEwan, Ryan)
Date: Tue, 21 May 2002 14:51:27 -0700
Subject: Summary:: Netra T1 AC200 anomoly
Message-ID: <AA217E43BBECC647A72A046243D7CF8937CD25@mxca1.netopia.com>

Turns out to be a bad cdrom was giving different hardware paths.  Crazy
thing was that with diagnostics turned on, it appeared that the cdrom was
being found, however using probe-ide, it was giving inconsistent errors.
Replaced cdrom and all is well again.

Many thanks to those that helped with suggestions.  Too many to list here.

Ryan

-----Original Message-----
From: Ryan McEwan [mailto:rmcewan at netopia.com]
Sent: Tuesday, May 14, 2002 4:06 PM
To: Sun Managers
Subject: Netra T1 AC200 anomoly


Perhaps somebody can explain this.

I posted earlier today about a jumpstart problem.  Many thanks to those that
helped.  Now I have an interesting anomoly, that apparently I have had for
quite some time, however I am just noticing the problem since this is the
first time these machines have been jumpstarted.  

I have probably 50 Netra AC200's.  Most of them look like this

#uname -a
SunOS esite6 5.8 Generic_108528-12 sun4u sparc SUNW,UltraAX-i2

#format
Searching for disks...done


AVAILABLE DISK SELECTIONS:
       0. c1t0d0 <SUN18G cyl 7506 alt 2 hd 19 sec 248>
          /pci at 1f,0/pci at 1/scsi at 8/sd at 0,0
       1. c1t1d0 <SUN18G cyl 7506 alt 2 hd 19 sec 248>
          /pci at 1f,0/pci at 1/scsi at 8/sd at 1,0
Specify disk (enter its number):


I have 1 netra though, that looks like this.

#uname -a
SunOS esite5 5.8 Generic_108528-12 sun4u sparc SUNW,UltraAX-i2

#format
Searching for disks...done


AVAILABLE DISK SELECTIONS:
       0. c0t0d0 <SUN18G cyl 7506 alt 2 hd 19 sec 248>
          /pci at 1f,0/pci at 1/scsi at 8/sd at 0,0
       1. c0t1d0 <SUN18G cyl 7506 alt 2 hd 19 sec 248>
          /pci at 1f,0/pci at 1/scsi at 8/sd at 1,0
Specify disk (enter its number):


Can Anyone explain to my why on 99 percent of my netra's the hardware path
is c1t0d0 and c1t1d1, however on this one machine it is c0t0d0 and c0t1d1?
I am at a loss.  It would be nice to know if there is a way to force it to
be c1t1d0 and c1t1d1 so that things like jumpstart work without creating
custom configs, etc.  I have gone to great lengths to keep things similar an
d am worried that this is going to burn me down the road because I have
forgotten the difference???

Any insight would be appreciated.

Ryan

[demime 0.99c.7 removed an attachment of type application/ms-tnef]
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From ak94601 at yahoo.com  Tue May 21 19:19:44 2002
From: ak94601 at yahoo.com (Ath Kar)
Date: Tue, 21 May 2002 16:19:44 -0700 (PDT)
Subject: Summary: Solaris8-GCC 3.0.3 compiler
Message-ID: <20020521231944.60261.qmail@web14807.mail.yahoo.com>

Hi 
Thanks to Darren, Michel, Kris, Tim, Hocke, Evans,
Randy, Daniel, Stace, Alex, Mark, Dragon, Frabrice,
Giardian, McCaffity

Everybody pointed towards installing binutil package
to solve the problem.

Also some advised to use 2.95.3 on PRODUCTION Server
But to take advantage of 64-bit I have installed
GCC3.0.3, before it I installed "binutil" package.

That worked... I successfully installed Samba2.2.2 &
Top3.5beta version.

Thanks

Ath


> Hi 
> 
> I have Solaris 8 installed on FunFire880. Now I have
> installed GCC 3.0.3, m4.1.4, make.3.77. Now I am
> trying to compile Samba-2.2.2 & top-3.5beta12 on it.
> But I am getting following errors....
> 
> Is GCC 3.0.3 is ok for use or I have some other
> problem . in both compilation I am getting errors
> related to /usr/ccs/bin & it is already in PATH.
> whats
> wrong ? Any suggestions ?
> 
> 
> /usr/ccs/bin/as: "/var/tmp/ccS3Zktg.s", line 1849:
> error: unknown opcode ".subsection"
> /usr/ccs/bin/as: "/var/tmp/ccS3Zktg.s", line 1849:
> error: statement syntax
> /usr/ccs/bin/as: "/var/tmp/ccS3Zktg.s", line 1905:
> error: unknown opcode ".previous"
> /usr/ccs/bin/as: "/var/tmp/ccS3Zktg.s", line 1905:
> error: statement syntax
> make: *** [smbd/server.o] Error 1
> 
> 
> 
> Thanks in Adv
> 
> Ath
LAUNCH - Your Yahoo! Music Experience
http://launch.yahoo.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From vijay_chint at yahoo.com  Tue May 21 20:13:22 2002
From: vijay_chint at yahoo.com (Vijay ch)
Date: Tue, 21 May 2002 17:13:22 -0700 (PDT)
Subject: Summary: mmap issue
In-Reply-To: <20020519223043.A13972@hermes>
Message-ID: <20020522001322.22802.qmail@web21502.mail.yahoo.com>

Thanks to Dave and sysadmin at astro.su.se
Solution:
---------
There's no need for to to be fixed - you just need to
use fixed offsets
from whatever pointer mmap returns. eg

    unsigned char *base;

    base = (unsigned char *) mmap(........);
    *(base+100) = 'A'; /* write an A to byte 101 of
the file */
    msync(); # flush changes
Question:
--------
   I have a group of applications which loads same
datafile into memory and exchange data using that
datafile. 

 One way I can do is mmap to use some fixed address
(say 0x2FC000000) and using MAP_FIXED option in all my
applications. Problem is it may not be safe to given
our own address 
  
 Is there any other way in which i can acheive the
same thing with out giving fixed address i.e using
MAP_SHARED option
LAUNCH - Your Yahoo! Music Experience
http://launch.yahoo.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From lraghunath at Lucent.Com  Wed May 22 03:46:17 2002
From: lraghunath at Lucent.Com (RaghuNath L(Raghu))
Date: Wed, 22 May 2002 13:16:17 +0530
Subject: Summary: logging in vxfs
In-Reply-To: <20020521125436.89128.qmail@web11508.mail.yahoo.com>
References: <5.1.0.14.0.20020521131544.02900e90@intpmail.iidc.lucent.com>
Message-ID: <5.1.0.14.0.20020522131126.02904ba0@intpmail.iidc.lucent.com>

Hello Managers,
Thanks for all who replied .

Every one pointing to very less time required to do an fsck ,but it does 
need an fsck .
It's a journal file system  which will make log automatically.
Regards
Raghu

At 05:54 AM 5/21/2002 -0700, Jason Wood wrote:
>VxFS is inherantly a journelled file system and needs no special options.
>
>--- "RaghuNath L(Raghu)" <lraghunath at Lucent.Com> wrote:
> > Hello ,
> > Does any body know what equivalent mount option of logging in vxfs ?
> > I don't want do an fsck  in case of abrupt shutdown.
> > i will sumaarize .
> > Regards
> > _______________________________________________
> > sunmanagers mailing list
> > sunmanagers at sunmanagers.org
> > http://www.sunmanagers.org/mailman/listinfo/sunmanagers
>
>
>=====
>Pinky: What will we do tonight Brain?
>Brain: Same thing we do every night Pinky..try to install Windows NT!
>(Saw it the other day and thought it was funny)
>LAUNCH - Your Yahoo! Music Experience
>http://launch.yahoo.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From Slamet.Fadilah at ptsi.siemens.co.id  Wed May 22 04:06:48 2002
From: Slamet.Fadilah at ptsi.siemens.co.id (Slamet Ahmad Fadilah)
Date: Wed, 22 May 2002 15:06:48 +0700
Subject: SUMMARY: SBus Board Swap (Off-Topic)
Message-ID: <812B90CE994D8349B861338E3BE0232B245DC8@jktg201e>

Sorry for a late summary.

Thanks to Carsten Boeddicker and Justin Stringfellow for their time
responding my email.
(Un-)fortunately I just got a new E4500, so I never try to swap the SBus
board.

Anyway please find their idea :

1. The swap could create a trouble since the controllers depends on the
position in the machine, they get the names from the position of (and on!)
the board. The system might loose the configuration data (because i.e. the
Controllers don't know their disks anymore). Especcially when
there are SCSI-Controllers on the board you should avoid an exchange.

2. It depends what sort of boards they are, and what software you are using.
Disksuite needs paths to say the same, veritas does not. JNI FCE-1064 cards
require the WWNs to stay in the same place hardware-wise, Sun FC-100 cards
do not...

BR, slamet

> -----Original Message-----
> From:	Slamet Ahmad Fadilah [SMTP:Slamet.Fadilah at ptsi.siemens.co.id]
> Sent:	Monday, May 13, 2002 2:15 PM
> To:	'sunmanagers at sunmanagers.org'
> Subject:	SBus Board Swap (Off-Topic)
> 
> Dear Manager,
> 
> My machine SUN E4500 has 2 SBus I/O boards. 
> SBus1 on slot 3 and SBus2 on slot 5.
> Any idea/experience if the position of both boards exchanged.
> 
> I will summarize.
> 
> Best regards,
> slamet
> _______________________________________________
> sunmanagers mailing list
> sunmanagers at sunmanagers.org
> http://www.sunmanagers.org/mailman/listinfo/sunmanagers
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From list-sun.managers at spoiled.org  Wed May 22 08:10:05 2002
From: list-sun.managers at spoiled.org (Juri Haberland)
Date: Wed, 22 May 2002 12:10:05 +0000 (UTC)
Subject: SUMMARY: Can Solaris do source routing like Linux with advanced
  routing?
References: <news2mail-a91am4$5jp$1@babel.spoiled.org>
Message-ID: <news2mail-acg1qt$rj8$1@babel.spoiled.org>

Whoops, sorry, I just discovered that I forgot to send this summary before
I went on holidays...

So here we go. My question was:

> I'm looking for a possebility to do 'source routing'. What I mean with it is
> that I want to be able to route based on the source, not the destination
> address. Think of the following:
> A box with two interfaces, each has an IP address in a different subnet
> and clients from the internet connect through both interfaces. Now I 
> want to route the 'answers' back through the corresponding interface where
> the 'question' came in. So I will need two default routes, but with this
> Solaris does a kind of round robing.
> With Linux I can use the advanced routing feature and can set up different
> routing tables and assign traffic to these tables based on e.g. the source
> address.

I received three answers:
- Giles Gamon send a link to a commercial tool called DefaultRouter.
www.defaultrouter.com

- Buddy Lumpkin suggested to set ip_enable_group_ifs=0, but this is the
default now and if I understand it correctly it is just for alias interfaces
and not different interfaces. I tried it anyway (also setting it to '1')
but it didn't work.

- Casper Dik wrote:
> ipfilter actually does allow you to route deliberately using the source
> address; I have two internet connections at home and use that feature
> with the two rules at the start of my ipf.conf:
> 
> pass out quick on qe0 to qe1:<qe1-router> from <qe1-address> to any
> pass out quick on qe1 to qe0:<qe0-router> from <qe0-address> to any
> 
> I have two interfaces here, qe0 and qe1; in the first rule, when a packet
> is seen "on qe0" with the wrong address (from <qe*1*-address>), I send it
> "to qe1" but make sure to direct the packet at "qe1-router" (the
> default route for the qe1 interface).
> 
> And the second rules send packets that should hav ebeen from qe0 but
> appear to be on qe1 back to qe0's default route.

I tried this one (man, compiling IPFilter was not the easiest task), and 
had partial success. When trying to connect to the second interface a
already established connection to the first interface would stall, but
the packets actually went out on the second interface.

Hmm, as it wasn't urgent and I didn't have a test box at that time I stopped
trying and decided to retry it if the test box gets available again.

Thanks everyone,
Juri

-- 
Juri Haberland  <juri at koschikode.com> 
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From akrasanth at velankani.com  Wed May 22 10:15:42 2002
From: akrasanth at velankani.com (AK Rasanth)
Date: Wed, 22 May 2002 19:45:42 +0530
Subject: Summary:time since epoch in secs
Message-ID: <02052219454266.01176@rasanth>

Hi  ,
Thanks to all who had replied to my posting.
the best answer I got to find time since epoch is :

/usr/bin/truss /usr/bin/date 2>&1 | /usr/bin/awk '/^time/ {print $NF}'
Thanks a lot Mr Dennis Patterson.
the other option is :

perl -e 'print time;'
Thanks to:

Dan Kelly
Schmitt Martin
Darren Moulding
Dave
Francisco
Ed.rolison
Bertrand Hutin
 
A.K.Rasanth
ph: 91-80-8522507 Ext 1626

--------------------------------------------------------------
Velankani Information Systems Ltd, Bangalore, India
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From deb at tickleme.llnl.gov  Wed May 22 11:26:57 2002
From: deb at tickleme.llnl.gov (Deb)
Date: Wed, 22 May 2002 08:26:57 -0700 (PDT)
Subject: SUMMARY- patches - a minor point to query
Message-ID: <200205221526.g4MFQvm2001272@tickleme.llnl.gov>

Many, many, MANY of you replied - all with the same answer:

Yes, a later patch always supercedes the earlier one, and 
contains everything therein, plus the new stuff.

There's so many replies, I can't possibly thank you directly,
but you have clarified this point for me quite well.

Thanks!

deb


=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-

"Press any key... no, no, no, NOT THAT ONE!!!"

t?t
 ~ 
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From bowens at eastman.com  Wed May 22 15:27:18 2002
From: bowens at eastman.com (Owens, Blaine C - Eastman)
Date: Wed, 22 May 2002 15:27:18 -0400
Subject: SUMMARY: POP3 server for Solaris
Message-ID: <D5F9E5AF86D815409CBB2AC499D57C6C068B3ED7@NTMAIL20A.emn.com>

The votes are in and the package getting the most votes was:
 
qpopper
 
A number of other packages got good reviews:
 
UWash imap server
cucipop
popa3d
cyrus
courierimap
gnu-pop3d
 
Thanks to all who responded! I'm going to be evaluating several of these
before I make my final decision.

Blaine Owens
Eastman Chemical Company
Phone - (423)-229-3579
Cell Phone - (423)-817-0704
Fax - (423)-229-1188
bowens at eastman.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From rfransix at yahoo.com  Wed May 22 16:57:05 2002
From: rfransix at yahoo.com (pelicancomputers)
Date: Wed, 22 May 2002 13:57:05 -0700 (PDT)
Subject: SUMMARY: loghost explained
In-Reply-To: <20020521144412.97486.qmail@web14803.mail.yahoo.com>
Message-ID: <20020522205705.42026.qmail@web14808.mail.yahoo.com>

thank you. everyone said the same thing, and here's a
good answer:

I can concurr - if there is an unexpected network
failure (or someone 
starts playing with the 'ifconfig hme0 unplumb'
command) you still get 
log files.

the only real reason to have loghost after a public IP
is if you are using a centralized log server (which I
would also recommend for those total system failures)

lastly (but certainly not leastly) I've configured my
systems to pass a single 'hosts' file between multiple
systems, and having 'loghost' under the loopback IP
means not having to recreate the hosts file for each 
system.

--- pelicancomputers <rfransix at yahoo.com> wrote:
> i read there is an advantage to having loghost in
> the /etc/hosts file, after 127.0.0.1 loopback entry:
> 
> 127.0.0.1     localhost    loghost    loopback
> 
> this prevents logging from stopping if the primary
> interface, where loghost is usually placed, stops
> working.
> 
> can you concur? are there other advantages?
LAUNCH - Your Yahoo! Music Experience
http://launch.yahoo.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From foster at dim.ucsd.edu  Wed May 22 17:43:41 2002
From: foster at dim.ucsd.edu (David Foster)
Date: Wed, 22 May 2002 14:43:41 -0700 (PDT)
Subject: [SUMMARY] Vulnerabilities in solaris in.rarpd (can I disable?)
Message-ID: <200205222143.g4MLhfw12987@dim.ucsd.edu>

Wanted to know when in.rarpd can be turned off, since a recent
advisory claiming local and remote vulnerabilities was posted.
My original question and the advisory follow.

in.rarpd is the daemon which determines IP addresses based on MAC 
addresses.

I only have two systems running in.rarpd, one is a SunRay server (hence
also a DHCP server) and the other is a Jumpstart server.

in.rarpd is required for boot servers (eg. Jumpstart or network boot), 
and if /tftpboot exists it will be started (see /etc/init.d/nfs.server). 
Many fire it up when installing servers via Jumpstart, then kill it.

Three folks claim that a DHCP server would require rarp, and I would tend
to agree; one claimed it would not.

Note that in.rarpd is a daemon, it is not started by inetd as a few
suggested. To disable it you must ensure /tftpboot doesn't exist so
it does not get started in /etc/init.d/nfs.server (or modify that
script). This disables bootp and therefore Jumpstart and (I believe)
DHCP as well.

I'll include Casper Dik's reply in it's entirety:

   My personal belief after examination of the code is this:

	- the routines in question are buggy
	- the routines in question are *always* called with fixed
	  parameters and exploitig them does not appear to be possible.

   Furthermore, rarp requests cannot pass routers and therefore
   the "remote" in the exploit is on directly attached networks.

   RARP is only used for network based installs (boot net - install)
   and other network booting (diskless)

   Casper
   
   
Thanks to:

topher
Tim Evans
Casper Dik
Michael Schulte
Jeff Kennedy
Jay Lessert
Michael Hocke
Dragon King

> 
> Just saw this advisory for in.rarpd, claims of local and remote
> vulnerabilities.
> 
> So my question is, can this be disabled? The daemon is only
> running on two of our systems, one is a SunRay server and the
> other is our NIS master, and I can't afford to break either.
> A preliminary search for information hasn't yielded anything useful, 
> and I'd like to resolve this asap if you know what I mean.
> 
> Thanks.
> 
> Dave Foster
> 
> > Mailing-List: contact bugtraq-help at securityfocus.com; run by ezmlm
> > X-Originating-IP: [144.134.252.220]
> > From: "david evlis reign" <davidreign at hotmail.com>
> > To: bugtraq at securityfocus.com, vulnwatch at vulnwatch.org
> > Subject: [DER Adv #7] - Multiple Vulnerabilities in solaris in.rarpd
> > Date: Wed, 22 May 2002 02:06:43 +0000
> > Mime-Version: 1.0
> > 
> > Intro:
> > rarpd is a reverse arp protocol for small to medium sized networks.
> > in the solaris implementation (in.rarpd) there seems to be 3 remotely 
> > exploitable buffer overflows, 2 locally exploitable and 2 cases of format 
> > string exploitability.
> > 
> > Details:
> > In the functions error and syserr (syserr also being used by other in.* 
> > implmentations which are also exploitable, but not the topic of this 
> > advisory today) there contains 2 common syslog calls without format strings.
> > 
> > static void
> > syserr(s)
> > char    *s;
> > {
> >         char buf[256];
> > 
> >         (void) sprintf(buf, "%s: %s", s, strerror(errno));
> >         (void) fprintf(stderr, "%s:  %s\n", cmdname, buf);
> >         syslog(LOG_ERR, buf);
> >         exit(1);
> > }
> > 
> > /* VARARGS1 */
> > static void
> > error(char *fmt, ...)
> > {
> >         char buf[256];
> >         va_list ap;
> > 
> >         va_start(ap, fmt);
> >         (void) vsprintf(buf, fmt, ap);
> >         va_end(ap);
> >         (void) fprintf(stderr, "%s:  %s\n", cmdname, buf);
> >         syslog(LOG_ERR, buf);
> >         exit(1);
> > }
> > 
> > there are two vulnerable calls which could be exploited locally or remotely.
> > 
> > vendor notification: nope
> > 
> > a working exploit has been created for the remote buffer overflows but not 
> > this time, not here.
> > 
> > DER systems
>    << All opinions expressed are mine, not the University's >>
> 
>   =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
>    David Foster    National Center for Microscopy and Imaging Research
>     Programmer/Analyst     University of California, San Diego
>     dfoster at ucsd.edu       Department of Neuroscience, Mail 0608
>     (858) 534-7968         http://ncmir.ucsd.edu/
>   =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
> 
>    "The reasonable man adapts himself to the world; the unreasonable one
>    persists in trying to adapt the world to himself.  Therefore, all progress
>    depends on the unreasonable."   -- George Bernard Shaw
> _______________________________________________
> sunmanagers mailing list
> sunmanagers at sunmanagers.org
> http://www.sunmanagers.org/mailman/listinfo/sunmanagers


   << All opinions expressed are mine, not the University's >>

  =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
   David Foster    National Center for Microscopy and Imaging Research
    Programmer/Analyst     University of California, San Diego
    dfoster at ucsd.edu       Department of Neuroscience, Mail 0608
    (858) 534-7968         http://ncmir.ucsd.edu/
  =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=

   "The reasonable man adapts himself to the world; the unreasonable one
   persists in trying to adapt the world to himself.  Therefore, all progress
   depends on the unreasonable."   -- George Bernard Shaw
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From granzowd at mail.nih.gov  Wed May 22 19:56:53 2002
From: granzowd at mail.nih.gov (Granzow, Doug (NCI))
Date: Wed, 22 May 2002 19:56:53 -0400
Subject: SUMMARY: How to mirror a failing drive?
Message-ID: <59445348FF4CD41182CF00508B6F779C06DA70BA@nihexchange11.nih.gov>

Thanks to:

	Guy B. Purcell
	Kevin Boykin
	Johnny Hall
	Darren Moulding

Several folks suggested using dd to copy the data to a partition on another
drive, instead of trying to get disksuite to mirror the data.  Unfortunately
the problematic drive failed completely before I was able to try this.
Fortunately, though, I was able to get a backup job completed just an hour
before the drive failed completely.  (We do regular backups but it is nice
to be able to restore from something so recent.)  I am now waiting for a
restore job to complete.

Thanks again for your input!

Doug



> -----Original Message-----
> From: Granzow, Doug (NCI) 
> Sent: Wednesday, May 22, 2002 11:55 AM
> To: Sun Managers List
> Subject: How to mirror a failing drive?
> 
> 
> I have a Disksuite 4.2 metadevice that, due to a string of 
> events, contains
> only one submirror.  One of the stripes on that mirror is now 
> marked as
> "Last Erred" in disksuite.  It is still online but I 
> occasionally get errors
> like the following in /var/adm/messages:
> 
> May 22 11:18:20 server      Error for Command: read           
>          Error
> Level: Fatal
> May 22 11:18:20 server unix:        Requested Block: 1757020
> Error Block: 1757117
> May 22 11:18:20 server unix:        Vendor: FUJITSU
> Serial Number: 0019530620  
> May 22 11:18:20 server unix:        Sense Key: Media Error
> May 22 11:18:20 server unix:        ASC: 0x13 (address mark 
> not found for
> data field), ASCQ: 0x0, FRU: 0x0
> May 22 11:18:20 server unix: WARNING: md: d52: read error on
> /dev/dsk/c0t12d0s6
> 
> When I try to attach another concat/stripe as a new 
> submirror, it starts to
> resync, but eventually gives up and leaves the new submirror 
> in "Attaching"
> status -- it never makes it to "OK".  Is there any way to get 
> Disksuite to
> mirror what it *can* read, so that when I replace this disk, 
> I do not lose
> *all* of the data?
> 
> Thanks,
> Doug
> _______________________________________________
> sunmanagers mailing list
> sunmanagers at sunmanagers.org
> http://www.sunmanagers.org/mailman/listinfo/sunmanagers
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From brounb at adi-limited.com  Thu May 23 01:14:56 2002
From: brounb at adi-limited.com (Broun, Bevan)
Date: Thu, 23 May 2002 15:14:56 +1000
Subject: Summary: SunFire-280R, L9, Networker: kern.notice
  lus_intr(3.0.0): transport failure
In-Reply-To: <20020522103821.K15205@adi-limited.com>; from
  brounb@adi-limited.com on Wed, May 22, 2002 at 10:38:21AM +1000
References: <20020522103821.K15205@adi-limited.com>
Message-ID: <20020523151455.A13565@adi-limited.com>

"Scott Buecker " <sbuecker at netjets.com> offered a suggestion but the
problem was that the L9 tape library was set to "sequential access" rather
than "random access". Basically this stopped networker being able to
control the auto changer.

We had commercial support for networker and I did get some good help from
this. In particular the "/etc/LGTOuscsi/sjirjc" command which I knew
nothing about.

root at host>/etc/LGTOuscsi/sjirjc 4.0.0   

                Device: 4.0.0

                   Number of Drives: 1
                 Number Drive Pairs: 1
   Number of Import/Export Elements: 0
      Number of Import/Export Pairs: 1
                    Number of Slots: 9
               Number of Slot Pairs: 1
       Number of Transport Elements: 1
          Number of Transport Pairs: 1

      Volume Tags Supported
Initialize Element Status Supported
        Bar Codes Supported        


when things were good. 

root at atlantis>/etc/LGTOuscsi/sjirjc 4.0.0
SJIRJC: Device busy 
Code:0x29, Str=<I/O error>

when things were bad.

<quote>
the sjirjc command doesn't use the jukebox driver: it directly queries the
jukebox robotic controller across the scsi bus to extract information. If
</quote>

anther command used was /etc/LGTOuscsi/inquire

BB

on Wed, May 22, 2002 at 10:38:21AM +1000, Broun, Bevan <brounb at adi-limited.com> wrote:
> I have a fresh Install of Solaris-8 with latest recommended patches. An L9
> tape library is being used by Networker version 5.5.4. On the first full
> backup networker had no problem with with the library. Today networker is
> having trouble operating the L9. The following is appearing in
> /var/adm/messages:
> 
> May 22 09:36:04 atlantis root: [ID 702911 daemon.notice] NetWorker media:
>     (warning) /dev/rmt/1cbn opening: I/O error
> May 22 09:42:30 atlantis last message repeated 3 times
> May 22 09:42:37 atlantis lus: [ID 497811 kern.notice] NOTICE:
>     lus_intr(3.0.0): transport failure (incomplete)
> May 22 09:43:14 atlantis last message repeated 3 times
> May 22 09:44:38 atlantis root: [ID 702911 daemon.notice] NetWorker media:
>     (warning) /dev/rmt/1cbn opening: I/O error
> 
> Looks like some patch is needed. Any info on "lus_intr" ?
> 
> TIA
> 
> BB
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From ral at codewan.com.ph  Thu May 23 05:23:46 2002
From: ral at codewan.com.ph (lito)
Date: 23 May 2002 17:23:46 +0800
Subject: SUMMARY:server resource monitoring
Message-ID: <1022145828.8711.103.camel@stn110>

Thank you to all.

The following are recommended by those who replied:

orcaware (http://www.orcaware.com/orca/)
setoolkit (http://www.setoolkit.com/)
TopItAll
rrdtool (http://people.ee.ethz.ch/~oetiker/webtools/rrdtool)
Net-SNMP 
sarge  (http://www.vais.net/~efinch/sarge.html)
ganglia (http://ganglia.sourceforge.net/)

of all, I think sarge and ganglia is the (at least suit my needs) best
of all, they're free but can do better than the commercial ones.

sarge rocks! very simple, straightforward, easy to install, exactly what
SME sysads will want. I prefer this one.

the best I think is ganglia, good clustering, portable, nice web
interface and best fitted for large nodes. A bit complicated than sarge,
but if you have volume of computers to monitor, this one will work well.

thanks again!    





Hi,
>
>I want to monitor my server resources status such as memory, CPU load,
>etc. I know top and iostat will do the work but I need utility program
>that can store history, generate a chart, identify the peak hour,etc
>(similar to webalizer) so I can produce a weekly and monthly report,
>preferrably web based. Does anyone use such kind of application for
>Solaris 2.6, if so, pls. help.  Thanks.
-- 
Lito A. Lampitoc		
Foundation for Communication Initiatives		phone:+63(2)8941345
CodeWAN Project					http://www.codewan.com.ph
--
Things which matter most must not be in the mercy of things which
 matter least" - Goethe.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From sun at bagdon.com  Thu May 23 06:13:13 2002
From: sun at bagdon.com (Steve Bagdon)
Date: Thu, 23 May 2002 06:13:13 -0400
Subject: SUMMARY - SUNWcsr missing pkginfo
Message-ID: <200205230613.AA368640294@mail.bagdon.com>

We've decided not to do anything about this, as Sun has declared any solution un-supportable (as Sun doesn't have a solution beyond reloading the OS), and we have a previous Sun case against this machine for this issue. We'll probably work around reloading the OS (it probably needs it anyway). But there were three general answers:

1) copy the file from another machine.

2) rebuild from a template, identifying the patches loaded for that package from 'showrev -p' and reading the info files from all the loaded (but not '-d' patchadd-ed patched) patches.

3) Thanks Asher:

1. boot cdrom -s 
2. mount / and /usr under /a and /a/usr 
2.5 cross fingers, knock on wood, etc 
3. pkgrm -M /a SUNWcsr 
4. pkgadd -M /a SUNWcsr 

Thanks to:

Chris Josephes
Sarah Eckhause
Michael Hocke
Kathryn Fielding
Asher Frenkel
Bertrand_Hutin
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From sajeev20 at rediffmail.com  Thu May 23 08:22:02 2002
From: sajeev20 at rediffmail.com (sajeev  nv)
Date: 23 May 2002 12:22:02 -0000
Subject: SUMMARY:EOD
Message-ID: <20020523122202.9320.qmail@webmail14.rediffmail.com>

An embedded and charset-unspecified text was scrubbed...
Name: not available
Url: http://www.sunmanagers.org/pipermail/summaries/attachments/20020523/bf517bfc/attachment.ksh 

From inzel at tunku.uady.mx  Thu May 23 11:14:26 2002
From: inzel at tunku.uady.mx (Israel Novelo)
Date: Thu, 23 May 2002 10:14:26 -0500
Subject: SUMMARY: POP3 server for Solaris (fwd)
In-Reply-To: <Pine.GSO.3.96.1020523092656.4072C-100000@apollo.sky.net>
Message-ID: <5.1.0.14.0.20020523101321.00bc5218@tunku.uady.mx>

Hi dear partners:

         i just Summary forwarding this message thanks a lot for all your 
help..

---------- Forwarded message ----------
>Date: Wed, 22 May 2002 15:27:18 -0400
>From: "Owens, Blaine C - Eastman" <bowens at eastman.com>
>To: sunmanagers at sunmanagers.org
>Subject: SUMMARY: POP3 server for Solaris
>
>The votes are in and the package getting the most votes was:
>
>qpopper
>
>A number of other packages got good reviews:
>
>UWash imap server
>cucipop
>popa3d
>cyrus
>courierimap
>gnu-pop3d
>
>Thanks to all who responded! I'm going to be evaluating several of these
>before I make my final decision.
>
>Blaine Owens
>Eastman Chemical Company
>Phone - (423)-229-3579
>Cell Phone - (423)-817-0704
>Fax - (423)-229-1188
>bowens at eastman.com
>_______________________________________________
>sunmanagers mailing list
>sunmanagers at sunmanagers.org
>http://www.sunmanagers.org/mailman/listinfo/sunmanagers
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From esanborn at snowshore.com  Thu May 23 11:10:26 2002
From: esanborn at snowshore.com (Ed Sanborn)
Date: Thu, 23 May 2002 11:10:26 -0400
Subject: SUMMARY: L1-A key for NetraX1
Message-ID: <200205231509.g4NF9BB06690@flyingfox.snowshore.com>

RE: L1-A key for NetraX1Hi folks,

  A  "Control-Break" works for Hyperterminal.


-Ed
  -----Original Message-----
  From: Francisco Puente (CEA) [mailto:francisco.puente at cea.ericsson.se]
  Sent: Thursday, May 23, 2002 10:05 AM
  To: Ed Sanborn
  Subject: RE: L1-A key for NetraX1


  Hi, i think you cannot break down the system from Hiperterminal, you should
try TeraTerm, from wich you can send a "Control-Break" at ANY time, getting
the ok prompt!

  I hope this help!
  Francisco Puente

  ERICSSON
  Francisco Puente
  Jambala Deployment Engineer
  Network Integration Center

  Compaqma Ericsson  S.A.C.I.
  *       Av. Eduardo Madero 1020- Piso 17
          C1106 ACX -  Buenos Aires - ARGENTINA
  *       Phone + 54 11 4319-5676
  *       Fax + 54 11 4316-0996
  *       e-mail: mailto:francisco.puente at cea.ericsson.se



  -----Original Message-----
  From: Ed Sanborn [mailto:esanborn at snowshore.com]
  Sent: Thursday, May 23, 2002 11:05 AM
  To: sunmanagers at sunmanagers.org
  Subject: L1-A key for NetraX1



  Hi folks,

     I have a PC with Hyperterminal hooked up to
  my LOM console port on my NetraX1.  When I apply
  power the system begins to boot. I would like to
  be able to break into the PROM monitor (the "OK"
  prompt).  Normally on a Sun keyboard you can hit
  the L1-A keys.  Any idea what the equivalent is
  from Hyperterminal?

  Thanks,

          Ed

  Ed Sanborn   (978) 367-8478
  Principal Network Administrator
  SnowShore Networks
  285 Billerica Rd.
  Chelmsford, MA. 01824
  _______________________________________________
  sunmanagers mailing list
  sunmanagers at sunmanagers.org
  http://www.sunmanagers.org/mailman/listinfo/sunmanagers
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From chendry at risstech.riss.net  Thu May 23 11:22:13 2002
From: chendry at risstech.riss.net (Chad Hendry)
Date: Thu, 23 May 2002 10:22:13 -0500
Subject: SUMMARY:  Sun 280R, 220R video
Message-ID: <NCBBICCOOLPKHICGADKFAENPGKAA.chendry@risstech.riss.net>

Original Post:
We had a four port Belkin Omniview switch and our 220R worked fine with it.
We sent it to another site and replaced it with an existing 8 port Omniview
Pro.    The 220R will not display on the switch and, in fact, the switch
does not detect the Sun's presence.  Another location uses a 16 port version
of the same switch and it works fine with their 280R.   We have a newer
Omniview Pro2.  We stretched a cable down to it and it displays.  In
addition the 4 port that worked with our 220R fails to work with the 280R at
the other location.  At one location, two 280R's with identical hardware are
attached to the same switch.  One works, one fails.

Does anyone have suggestions on configuring the video on the 220 and 280
systems to work with various switches and monitors?

SUMMARY:
It appears that our problem must be unique.  The only idea seemed to be one
we'd tried, with no success, as follows:

The only time I have ran into this problem with an omni view, I reset it.
Next, made sure it was on the # I was booting and let it go through the
whole boot process and it worked.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From dijon at ratcage.com  Tue May 21 18:21:34 2002
From: dijon at ratcage.com (Doug Krause)
Date: Tue, 21 May 2002 18:21:34 -0400
Subject: Summary:  Solaris Network Command
Message-ID: <3CEAC86E.20601@ratcage.com>

Thanks for all that responded so quickly to my question:

> I'm trying to remember the name of a command I've used under Solaris 
> before.  It lists the MAC addresses of all the computers that it can see 
> on the local network.  Any ideas?  Thanks.


The command I was looking for was "arp -a".  Alternately some people
said that "netstat -p" does the same thing.  The one thing to remember
is to ping the subnet you are on so that all the attached computers
are put in the arp table.

-- 
Doug Krause
dijon at ratcage.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From fan at cns.nyu.edu  Tue May 21 21:14:15 2002
From: fan at cns.nyu.edu (fan at cns.nyu.edu)
Date: Tue, 21 May 2002 21:14:15 -0400 (EDT)
Subject: SUMMARY: HELP: problems with mail clients
Message-ID: <200205220114.VAA05481@calaf.cns.nyu.edu>

Hi -

Before anyone waste any time on the problems described
below, it turns out the source the problems is that
lockd had died on our server box.

Best,
Paul

The original message:
---------------------------------------------------------------------
I have some mysterious problems related to mailx and pine.
The machine is a Solaris 5.8 box.

The following commands will just hang.

  mailx
  mailx -f foomail, where foomail is in a user's home directory
  cat foomail  (after mailx -f foomail has been invoked)    
  cp foomail foomail2 (after mailx -f foomail has been invoked)

pine will hang if it is accessing mail folders in a user's home directory
and it will hang when it tries to send out mail.

User's home directories are NFS filesystems.

However, mailx -f /var/mail/logname will work and if foomail is moved
to /tmp, then mailx -f foomail will also work.
-------------------------------------------------------------------------
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From Fisher at cc.admin.unt.edu  Thu May 23 12:48:57 2002
From: Fisher at cc.admin.unt.edu (Nancy Fisher)
Date: Thu, 23 May 2002 11:48:57 -0500
Subject: SUMMARY How to set up non-root user to run shutdown
Message-ID: <scecd72c.012@gwia.unt.edu>

Hello,
I want to allow a non-root user to run /etc/shutdown command.  I am
running Solaris 8 2/2 and the username is the owner with group as
staff.

I entered the following:

setfacl -m user:username:5 /etc/shutdown

getfacl /etc/shutdown showed:
#owner:root
#group:sys
user::rwx
user:username:r-x    #effective:r-x
group:r-x                   #effective:r-x
mask:r-x
other:r-x

When username runs the /etc/shutdown, it says only root can shutdown.
What am I doing wrong?  Thanks in advance for responses. 
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From gio at uslink.net  Thu May 23 15:48:06 2002
From: gio at uslink.net (Giovanni Navarrette)
Date: Thu, 23 May 2002 14:48:06 -0500 (CDT)
Subject: SUMMARY: Bandwidth quota options?
Message-ID: <Pine.GSO.4.05.10205231443520.22972-100000@netapp>

Hey everyone:

Here's the ideas I got. Its basically what I guessed, it will have to be a
function of the webserver. We're running Zeus (I'd prefer apache..., but
what do I know :) ) so I'll have to find a module for that that works :D

Option1: apache mod_throttle and mod_bandwidth
Option2: use a linux server with CBQ (Class based queueing)
Option3: Sun bandwidth module? (Supposidly they make one)
Option4: Dyband software makes software for policing bandwidth to and
from a box. (http://www.dyband.com)

Thanks everyone!

-------------------------------------
Giovanni Navarrette
USLink Internet Systems Administrator
e-Mail: gio at uslink.net
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From robert.legate at telus.com  Thu May 23 17:15:42 2002
From: robert.legate at telus.com (Robert Legate)
Date: Thu, 23 May 2002 15:15:42 -0600
Subject: SUMMARY: 280r won't boot from cdrom
Message-ID: <23D5B89EC7B4D311821400805F85082C04C21C61@TAC_NT6>

Thanks all, 

We tried reseating all components and still nothing. Called in SUN, who had
brought a replacement dvd drive just in case, put in an original 10/01 cd
and away it went. It may be my cd burner is making faulty cd's, am testing
that now.  We did try an original 10/01 yesterday and it wouldn't work, but
we we're unable to duplicate the problem when the Sun guys were here. Go
figure.

>  -----Original Message-----
> From: 	Robert Legate  
> Sent:	Wednesday, May 22, 2002 3:37 PM
> To:	'sunmanagers at sunmanagers.org.'
> Subject:	280r
> 
> I have a brand-new out-of-the-box 280r with 1 cpu and 1GB RAM and upon
> trying to boot off the Solaris 8 10/1 install cd I get the following:
> 
> ok boot -s cdrom
> Boot device: /pci at 8,600000/SUNW,qlc at 4/fp at 0,0/disk at 0,0  File and args: -s
> cdrom
> krtld: load_exec: fail to expand cpu/$CPU
> krtld: error during initial load/link phase
> panic - boot: exitto64 returned from client program
> Program terminated
> 
> and after having set the boot device order at the BPROM level to boot off
> the cdrom first:
> 
> 
> ok boot -s
> Boot device: /pci at 8,700000/scsi at 6/disk at 6,0:f  File and args: -s
> 
> CPU seeprom format: 0000.0000.0000.0002
> screen not found.
> keyboard not found.
> Keyboard not present.  Using ttya for input and output. .......
> 
> 
> Has anyone run across this before?
> 
> Will summarize,
> thx
> 
>  <<...OLE_Obj...>> 
> 
> G. Robert Legate
> Network Operations Support Systems
> 780-493-2625 office
> 780-498-6540 cell
> robert.legate at telus.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From rnf at tbird.com  Thu May 23 17:33:32 2002
From: rnf at tbird.com (Rick Fincher)
Date: Thu, 23 May 2002 17:33:32 -0400
Subject: Summary: Solaris 9 End of Life?
References: <8F00B448EF423D4693744453E73FC90F24E00B@sm-cala-xm01.swna.wdpr.disney.com>
Message-ID: <012201c202a1$7cb28660$117a48c7@rickspc>

The short answer is Solaris 9 supports sun4m but may not in future releases,
and drops support for sun4d.

Rick

> Check out Sun's website:
>
>
http://docs.sun.com/ab2/coll.214.12/S9RELNOTESUPP/@Ab2PageView/683?Ab2Lang=C
&Ab2Enc=iso-8859-1
>
>
> If I read it right it say the 10's and 20's are supported now but may not
be in future releases.
>
>
> End-of-Support Products
> Products Not Supported in the Solaris 9 Operating Environment
> Support for the following products has been discontinued. For more
information, contact your support provider.
>
> sun4d Servers
> The following sun4d architecture servers are no longer included in this
release:
>
> SPARCserver(TM) 1000 systems
>
> SPARCcenter(TM) 2000 systems
>
>
> Hardware options that are dependent on the sun4d architecture are no
longer included in this release.
>
> Ethernet Quad Drivers qe and qec
> Ethernet Quad drivers qe and qec are no longer included in this release.
>
> Alternate Pathing Multipath I/O
> Alternate Pathing (AP) multipath I/O technology is no longer included in
this release. It has been retired in favor of the newer, more scalable
technologies presented by MPxIO and IPMP.
>
> These modern technologies present better overall multipath solutions, with
refined, user-friendly interfaces that are well integrated with Solaris.
IPMP also provides true automatic switching for networks upon error
detection.
>
> If you used AP in previous releases of Solaris for I/O multipath
capabilities, you are encouraged to utilize these newer technologies for I/O
multipath control.
>
> Future End-of-Support Products
> Support for the following products may be discontinued in future releases.
For more information, contact your service provider.
>
> PC File Viewer
> PC File Viewer may no longer be supported in a future release.
>
> ShowMe TV
> ShowMe TV may no longer be supported in a future release.
>
> FDDI, HSI/S, le, and Token Ring Drivers
> FDDI/S, FDDI/P, HSI/S, le, SBus Token Ring, and PCI bus Token Ring network
drivers may no longer be supported in a future release.
>
> SPC Driver
> The SPC S-Bus interface card driver may no longer be supported in a future
release. The SPC card is a Serial Parallel Controller (S-Bus, 8-port serial,
1 parallel).
>
> Sun4m Platform Group
> The sun4m platform group may no longer be supported in a future release.
This would affect the following platforms:
>
> SPARCclassic
>
> SPARCstation LX / LX+
>
> SPARCstation 4
>
> SPARCstation 5
>
> SPARCstation 10
>
> SPARCstation 20
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From Srinivas_Arella at Satyam.com  Fri May 24 00:52:02 2002
From: Srinivas_Arella at Satyam.com (Srinivas_Arella at Satyam.com)
Date: Fri, 24 May 2002 10:22:02 +0530
Subject: Summary: E250 hanging at boot
Message-ID: <7FF62A49079FD511B14400065B19EF1205C0BF@cpr.satyam.com>

the problem turned out to be a failed HDD. Though there was nothing in the
vfstab and it is not a part of the md device, it is causing the solaris to
wait for a very long time before booting. After removing the faulty disk the
system boots like charm...I am still confused about how could a failed SCSI
cause the system to give a long pause. Anyone has clues?

Rene Occelli's mail helped alot. Here are others who gave helpful
hints..thanx to all of ya..:)

Rahul Parasnis [rahul.parasnis at creditlyonnais.fr]
Matt D. Harris [mdh at mdh.si.edu]
Kai Riasol Gonzalez [riasol at web.de]
Dennis Martens [Dennis_Martens at health.qld.gov.au]

-----Original Message-----
From: Srinivas_Arella at satyam.com [mailto:Srinivas_Arella at satyam.com]
Sent: Tuesday, May 21, 2002 1:14 PM
To: sunmanagers at sunmanagers.org
Subject: E250 hanging at boot


Hi All,

we have just received an E250 from another place(which obviously was
operational). The machine just refuses to boot even on cdrom. All
diagnostics come out clean. I noticed that one out of the six disk slots in
empty ( take out??!!! ). The missing disk apears to be on disk6 ( and
presumably not a boot disk).

The machine seems to boot fine till printing the Sun OS header and copyright
information ( on CD or from boot disk). I am unable to contact the admin who
handled the machine previously. Any help in tracing the problem would be
great.


TIA,
Srinivas
************************************************************************** 
This email (including any attachments) is intended for the sole use of the
intended recipient/s and may contain material that is CONFIDENTIAL AND
PRIVATE COMPANY INFORMATION. Any review or reliance by others or copying or
distribution or forwarding of any or all of the contents in this message is
STRICTLY PROHIBITED. If you are not the intended recipient, please contact
the sender by email and delete all copies; your cooperation in this regard
is appreciated.
**************************************************************************
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers
************************************************************************** 
This email (including any attachments) is intended for the sole use of the
intended recipient/s and may contain material that is CONFIDENTIAL AND
PRIVATE COMPANY INFORMATION. Any review or reliance by others or copying or
distribution or forwarding of any or all of the contents in this message is
STRICTLY PROHIBITED. If you are not the intended recipient, please contact
the sender by email and delete all copies; your cooperation in this regard
is appreciated.
**************************************************************************
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From jhl22 at yahoo.com  Fri May 24 10:26:47 2002
From: jhl22 at yahoo.com (Jeff Lucas)
Date: Fri, 24 May 2002 07:26:47 -0700 (PDT)
Subject: SUMMARY: How to script labeling of new disks to avoid manually
  using format?
Message-ID: <20020524142647.9591.qmail@web11003.mail.yahoo.com>

Wow. I did not expect to get bombarded with responses!

Thanks to all!

In summary, there were many variations of piping
prtvtoc output to fmthard.  Some examples are as
follows...

##############
Example 1
##############
label one disk, let's say /dev/rdsk/c0t0d0s2, then do
prtvtoc /dev/rdsk/c0t0d0s2 | fmthard -s - <every other
disk>

##############
Example 2
##############
for i in 1 2 3 4 5 6 7 8 9 ; do
prtvtoc /dev/rdsk/c1t0d0s2 | fmthard -s -
/dev/rdsk/c2t0d$is2
done

There were other suggestions which included using
perl.  Let me know if you want them.

Thanks.

-Jeff
LAUNCH - Your Yahoo! Music Experience
http://launch.yahoo.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From min at psde.mec.mei.co.jp  Fri May 24 11:16:33 2002
From: min at psde.mec.mei.co.jp (Min Oo Tint)
Date: Fri, 24 May 2002 16:16:33 +0100
Subject: SUMMARY: command get serial number of your sun
Message-ID: <3C667B62E467F44DB51E01DC328A34311362D4@psde-server>

Hi,

thanks for so many replies.

Simple answer is no!

some suggest it hostid which I know is used by lmgrd and is not S/N

and 

eeprom | grep serial

Good suggestion I have is to read the s/n from the back of the machine and
write it down on and save it in /etc/systemserialnumber

 and one funny reply was:

yeah, something like:
"Hey, you, junior! Get down on your knees at the back of the server and
scribble
down the almost indecipherable number on a scrap of paper."

> We have several suns and would like a script that would gather all the
> serial number and produce a report.


Oh, you mean software! No way, the serial number is only on the case

--


and I did search the old achieve and found that it is not possible but hope
I would get yes.

http://www.sunmanagers.org/pipermail/summaries/2001-July/001013.html


-----Original Message-----
From: Min Oo Tint 
Sent: Friday, May 24, 2002 1:23 PM
To: Sun Managers (E-mail)
Subject: command get serial number of your sun
Importance: High


Hi All,

Does a command exists to get the serial number of your sun Solaris machine.
We have several suns and would like a script that would gather all the
serial number 
and produce a report.

Cheers,

Min Oo

Panasonic System LSI Design Europe
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From jeffw at pbs.com  Fri May 24 11:19:31 2002
From: jeffw at pbs.com (Jeff Wellens)
Date: Fri, 24 May 2002 10:19:31 -0500
Subject: SUMMARY: swap device truncated message
Message-ID: <NEEBKGEBEINHGKGIGFLCEEEECJAA.jeffw@pbs.com>

This message is just stating that my swap space is over 2gig in size and
that it will only be using 2gig and not the rest.  It is harmless.

Thanks so much for all the responded to my questions.

---------------------------------------------------------

Original question:

I am getting this error when rebooting new system just installed with
Solaris 7.  We are having some strang database issues and I am wondering if
this error may have something to do with it.  Does anyway know what this
means or if I should be concerned about it?

May 22 07:44:15 LaPorte unix: NOTICE: swap device /dev/md/dsk/d20 truncated
from  80796000 to 7fffffff


Thanks.
 Jeff Wellens
Publishing Business Systems
Technical Support Analyst
651-634-9218
jeffw at pbs.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From ypismerov at tucows.com  Fri May 24 11:45:28 2002
From: ypismerov at tucows.com (Yura Pismerov)
Date: Fri, 24 May 2002 11:45:28 -0400
Subject: SUMMARY: partitioning for Oracle
References: <7D710C9705F3D311953B00D0B73ECCBB04865E17@tpaexch.ariba.com>
Message-ID: <3CEE6018.8C7ED5CB@tucows.com>

Heflin Hogan wrote:
> 
> There is no real reason to allocate a large number of partitions for an
> oracle instance anymore, and even Oracle doesn't recommend doing things taht
> way anymore. The idea was to spread IO out over several drives and channels
> to improve performance, but it doesn't really apply with modern storage
> subsystems (unless you can convince your management to buy several T3 arrays
> to cover the DBAs' request). Oracle has a white paper out on something
> called the Oracle Flexible Architecture (I think), which covers the way
> Oracle thinks the database directory structure should be laid out, but it
> doesn't specify a physical allocation method.
> 
> -Heflin Hogan
> Systems Integrator
> Ariba, Inc.

	Thanks for the reply !

	Also big thanks to:

Jonathan Burelbach 
Lin Feng
Simon Greenland
Erwin Fritz
Jay Koonz
Steve Hunt
Alex Pleszko
Alexandre Perematko
John Phillips
Carl Gobbo
Steven Haywood
Andrew Rotramel
Michael Maciolek

Sorry if I don't mention somebody since the replies are still coming.

The conclusion is, the proposed partitioning method is obsolete and is
to be reviewed due to dramatic 
changes in storage technology field that happened since the original
Oracle recommendations were written.
There is still some reasons to create multiple partitions but having a
dozen of them per instance is overkill. The "rule of thumb" is keep
data, redologs, archived logs and control files on separate
partitions (better yet spindles or disk groups in case of RAID
configuration) in order to minimize
disk IO impact. In case of small configurations when everything resides
on one spindle it does not 
make much sense to create a separate partition for each class of Oracle
data. And of course it does not make sense at all to create multiple
partitions for the same type/class (ie data1, data2, data3, etc).
	
Thanks again everybody. This list rocks ! :)


> 
> -----Original Message-----
> From: Yura Pismerov [mailto:ypismerov at tucows.com]
> Sent: Friday, May 24, 2002 9:18 AM
> To: sunmanagers at sunmanagers.org
> Subject: partitioning for Oracle
> 
>         Sorry for off-topic,
> I'm trying to "fight" DBA regarding partitioning method they propose for
> our Oracle boxes.
> Every time they request a few dozen (!) partitions such as data[0-N],
> arch[0-N], redolog[0-N],
> trace[0-N] etc etc.
> 
> In my opinion it does not make sense since the mentioned partitions are
> located on
> RAID1+0 so nothing is gained in terms of performance. Also I have to
> mess around with moving data once
> one of the partitions gets full. Not to mention fewer number of
> partitions is obviously easier to
> maintain in such events as manual filesystem check.
> 
>         Could anybody tell me if there are real advantages of doing that
> crazy
> partitioning ?
> Our DBA manager is going to conduct "educational" session with
> operations team to explain the reasons
> they keep doing this. I just need some more arguments...
> 
> Thanks in advance.
> 
> --
> 
> Yuri Pismerov, Sr. System Administrator,
> TUCOWS.COM INC. (416) 535-0123  ext. 1352
> _______________________________________________
> sunmanagers mailing list
> sunmanagers at sunmanagers.org
> http://www.sunmanagers.org/mailman/listinfo/sunmanagers

-- 

Yuri Pismerov, Sr. System Administrator, 
TUCOWS.COM INC.	(416) 535-0123  ext. 1352
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From sunman at ureach.com  Fri May 24 12:34:26 2002
From: sunman at ureach.com (Vahid Moghaddasi)
Date: Fri, 24 May 2002 12:34:26 -0400
Subject: SUMMARY:SZ and RSS column in ucb/ps
Message-ID: <200205241634.MAA15666@www20.ureach.com>

Well, I have received may replies from admins and they all mentioned that my 
statement is not necessarily true, although I see this happens ONLY on our 
home-grown application. I have not not received any explanation as to why some of 
our applications have larger RSS that SZ or in general why is it possible to have 
RSS>SZ.
pmap/pmap -x was suggested by Casper Dik which prints lots of information about a 
running process, I found it very useful.
Many thanks to the following for reply:
Darren Dunham
Kim, Daniel J
Casper Dik
Urie, Todd

Vahid.
> Subject: SZ and RSS column in ucb/ps
> Date: Wed, 22 May 2002 14:25:33 -0400
> 
> 
> Hi Everyone,
> Is this a correct assumption that if RSS is larger than SZ, then the application 
have 
> memory leak?
> SZ=total size of the process
> RSS=resident set size of process
> 
> Thank you all.
> 

________________________________________________
Get your own "800" number
Voicemail, fax, email, and a lot more
http://www.ureach.com/reg/tag
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From santode at clcsmail.ksc.nasa.gov  Fri May 24 12:33:21 2002
From: santode at clcsmail.ksc.nasa.gov (Deborah Santomauro)
Date: Fri, 24 May 2002 12:33:21 -0400 (EDT)
Subject: SUMMARY: Problem with package creation......
Message-ID: <200205241633.g4OGXES16897@clcsmail.ksc.nasa.gov>

Thanks to:

Todd Urie
Serge Borkov



ANSWER:

Problem solved by including "cat > $1 <<EOM" within the 'checkinstall' script. 
This is what we were missing and once it was added, our problem was solved.



(ORIGINAL PROBLEM)

We've poured over the "Applications Packaging Developers Guide" and followed the 
instructions to the letter, but we're still having a problem.

In changing the "CLASSES" variable to a particular class name within the 
"checkinstall" script, we have found that the files for that class do not get 
installed. 

Furthermore, in the "preinstall" script and "postinstall" script, the "CLASSES" 
variable does not show the value it was set to in the "checkinstall" script.

Perhaps one of you has a working example that you can share????

We have tried this on both 2.6 and 8 and so far haven't found any answers in the 
Sun docs web site nor does there appear to be a bug report, patch, etc for 
pkgadd or pkgmk that may be a cause for the problem we're having.

Chimo,
-deb

-------------------------------------------------------------------------------
Deborah Santomauro
Unix System Administrator - SCSA
Kennedy Space Center - Florida / CSOC-CLCS-ROS
email: santode at clcsmail.ksc.nasa.gov 
Work: 321-861-2226				/\_/\
			       		       ( - - )
			    		     mm \ T / mm
-------------------------------------------------------------------------------
I don't know what I'm doing, but I'm doing it as hard as I possibly can.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From mikelist at sky.net  Fri May 24 12:42:48 2002
From: mikelist at sky.net (Mike's List)
Date: Fri, 24 May 2002 11:42:48 -0500 (CDT)
Subject: SUMMARY: /etc/mail/access file
In-Reply-To: <Pine.GSO.3.96.1020524102740.14043B-100000@apollo.sky.net>
Message-ID: <Pine.GSO.3.96.1020524114117.14649B-100000@apollo.sky.net>

Thanks to Dave and Lars...from /usr/lib/mail/README (which I didn't have)
DISCARD was the option I was looking for, low system resources used.


- Mike


The value part of the map can contain:

        OK              Accept mail even if other rules in the
                        running ruleset would reject it, for example,
                        if the domain name is unresolvable.
        RELAY           Accept mail addressed to the indicated domain or
                        received from the indicated domain for relaying
                        through your SMTP server.  RELAY also serves as
                        an implicit OK for the other checks.
        REJECT          Reject the sender or recipient with a general
                        purpose message.
        DISCARD         Discard the message completely using the
                        $#discard mailer.  If it is used in check_compat,
                        it affects only the designated recipient, not
                        the whole message as it does in all other cases.
                        This should only be used if really necessary.
        ### any text    where ### is an RFC 821 compliant error code and
                        "any text" is a message to return for the command.
                        The string should be quoted to avoid surprises,
                        e.g., sendmail may remove spaces otherwise.
        ERROR:### any text
                        as above, but useful to mark error messages as
such.
        ERROR:D.S.N:### any text
                        where D.S.N is an RFC 1893 compliant error code
                        and the rest as above.


On Fri, 24 May 2002, Mike's List wrote:

> Is it possible to put /dev/null in your /etc/mail/access file?
> 
> Just been to sendmail.org and I know you can use DISCARD to accept and
> discard incoming mail but is /dev/null works faster/allowable in access?
> 
> domain.com	DISCARD		(know it's possible)
> domain.com	/dev/null	(equivalent to above or possible?)
> 
> 
> 
> - Mike
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From Doug_S_Johnson at raytheon.com  Fri May 24 12:47:11 2002
From: Doug_S_Johnson at raytheon.com (Doug S Johnson)
Date: Fri, 24 May 2002 11:47:11 -0500
Subject: Summary: SunOS 4.1.4 Restore
Message-ID: <OFF8175FF5.80495140-ON86256BC3.005C2750@mck.us.ray.com>

Thanks to those who provided advice

You cannot simply mount the drive on a solaris machine and run ufsrestore 
because SunOS does not use ufs filesystems it uses 4.2 and uses the 
commands dump and restore instead of ufsdump and ufsrestore..I had already 
tried and ufsrestore couldn't read the tape and I don't have another SunOS 
machine.

I borrowed a Solaris 1.1.2 CD (I'll make a copy) and after that everything 
went smoothly. Below is a condensed version of what happens

ok> boot cdrom

What would you like to do?
  1 - Install SunOS miniroot?
  2 - Exit to single user shell
Enter a 1 or 2: 1

Begins a probe for disks and you select your root disk from a list.

Do you want to format and or label disk "sd0"?
  1 - yes, run format
  2 - no, continue with loading miniroot
  3 - no, exit to single user shell
Enter a 1,2,or 3: 2

Miniroot gets installed to /dev/sd0b (swap)

What would you like to do?
  1 - reboot using the just installed miniroot
  2 - exit into single user shell
Enter a 1 or 2: 1

System boots to miniroot

# mount /dev/sd0a /a

Run restore from here..(I won't bore you with any details from here, 
restore works like ufsrestore).
Commands you might need to use are located in the etc directory of the 
miniroot.

# reboot
_________________________________________
Original post below:

        I have a Sparc 20 with SunOS 4.1.4, it's rarely used but just when
it's needed something happened to it. The /dev directory was removed (I
didin't do it, but I need to fix).  Anyway, I know how to fix with
solaris: boot cdrom -s, mount c0t0d0sX /mnt, cd /mnt, ufsrestore from
tape.  However, I don't know how to in SunOS.

        I've read about booting from cdrom and tape, installing miniroot,
but I haven't seen any how to on it.  I believe you have to boot off a
tape or cdrom and somehow install miniroot to /tmp and then boot to the
miniroot, mount the filesystem,

        .  I'm borrowing a  SunOS 4.1.3 cd tomorrow (Can't find 4.1.4),
but don't know the steps to boot to miniroot.  I have backup tapes. How is
this done?

        Thanks, for any help you can provide.

 DJ
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From vijay_chint at yahoo.com  Sat May 25 01:44:15 2002
From: vijay_chint at yahoo.com (Vijay ch)
Date: Fri, 24 May 2002 22:44:15 -0700 (PDT)
Subject: Summary: how to get process status using pid
In-Reply-To: <OFCFBC7BB1.C255336A-ON42256BC3.00274DE5@com>
Message-ID: <20020525054415.14047.qmail@web21509.mail.yahoo.com>

My question:
How to get the process status (Running/Stopped)
using pid, using C in solaris.

Solution:
Process's info is stored into psinfo of /proc/<pid>/
and its structure 
is
defined in /usr/include/sys/procfs.h (typedefed
psinfo_t)
If you want to know process's status (ie. R,S,W
etc...) you should 
examine
pr_flag of psinfo_t structure. its type is int and its
mean is also 
defined
in same file. 

Thanks zoo11 at mail.hst.co.kr and Karl Vogel for
answering my question
Thanks Johan Hartzenberg for your discovery

Please find below the c code sent by Karl Vogel
-----------------------------------------------
---------------------------------------------------------------------------
/*
 * ftp://ftp.eng.auburn.edu/pub/doug/qps.c
 * A very fast ps that goes through procfs
 * @(#)qps.c	1.21	08/02/97		Doug Hughes Auburn
University
 */

#include <stdio.h>
#include <sys/types.h>
#include <sys/fcntl.h>
#include <sys/procfs.h>
#include <string.h>
#include <stdlib.h>
#include <pwd.h>
#include <dirent.h>
#include <sys/mkdev.h>

#ifndef FALSE
#define FALSE 0
#endif
#ifndef TRUE
#define TRUE 1
#endif

static int reverse=0;

typedef struct qps {
	int pid;
	char ppid[40];
	char tty[3];
	char user[9];
	float cpu;
	float mem;
	char pname[17];
	char *arglist;
} *QPS;
	
void usage() 
{
	puts("qps [-acdmnptruPU]");
	puts("  -a  show initial (80 chars of) process
args");
	puts("  -A  show ALL arguments");
	puts("  -c  sort by CPU usage");
	puts("  -d  debug (will print process ID before
opening process)");
	puts("  -e  display environment variables");
	puts("  -m  sort by memory usage");
	puts("  -n  sort by process name");
	puts("  -p  sort py pid");
	puts("  -P  show parent process and PGID");
	puts("  -t  sort by tty");
	puts("  -r  reverse order sorting");
	puts("  -s  memory size info SIZE and RSS in pages
and bytes 
(respectively)");
	puts("  -u  sort by user");
	puts("  -U  do not attempt to do uid->username
mapping (Save CPU)");
	exit(1);
}

int sort_by_pid(const void *farg1, const void  *farg2)
{
	QPS arg1, arg2;

	arg1 = (QPS) farg1;
	arg2 = (QPS) farg2;
	if (reverse)
		return(arg2->pid > arg1->pid);
	else
		return(arg1->pid > arg2->pid);
}

int sort_by_cpu(const void *farg1, const void *farg2)
{
	QPS arg1, arg2;

	arg1 = (QPS) farg1;
	arg2 = (QPS) farg2;
	if (reverse)
		return(arg2->cpu < arg1->cpu);
	else
		return(arg1->cpu < arg2->cpu);
}

int sort_by_mem(const void *farg1, const void *farg2)
{
	QPS arg1, arg2;

	arg1 = (QPS) farg1;
	arg2 = (QPS) farg2;
	if (reverse)
		return(arg2->mem < arg1->mem);
	else
		return(arg1->mem < arg2->mem);
}

int sort_by_tty(const void *farg1, const void *farg2)
{
	QPS arg1, arg2;

	arg1 = (QPS) farg1;
	arg2 = (QPS) farg2;
	if (reverse)
		return(strcmp(arg2->tty, arg1->tty));
	else
		return(strcmp(arg1->tty, arg2->tty));
}

int sort_by_user(const void *farg1, const void *farg2)
{
	QPS arg1, arg2;

	arg1 = (QPS) farg1;
	arg2 = (QPS) farg2;
	if (reverse)
		return(strcmp(arg2->user, arg1->user));
	else
		return(strcmp(arg1->user, arg2->user));
}

int sort_by_pname(const void *farg1, const void
*farg2)
{
	QPS arg1, arg2;

	arg1 = (QPS) farg1;
	arg2 = (QPS) farg2;
	if (reverse)
		return(strcmp(arg2->pname, arg1->pname));
	else
		return(strcmp(arg1->pname, arg2->pname));
}

/*
 * Subproc to display all args (as opposed to what's
in arglist
 */
char *showallargs(int fd, struct prpsinfo *p)
{
	/* routine rewritten by Roger Faulkner - procfs
co-designer and guru 
*/
	char **argv;					/* hold arg pointers */
	char *env[400];					/* hold environment */
	int argc = p->pr_argc;			/* arg count */
	char buf[BUFSIZ+1];				/* temporary buffer */
	int len;						/* strlen(buf) */
	char *retbuf= NULL;				/* to return arglist */
	int argsize=64;					/* keep argsize for allocation
records */
	int retbuflen;					/* strlen(retbuf) */
	int i = 0;
	char *sp;							/* a space character */


	if ((retbuf = malloc(argsize)) == NULL) {
		fprintf(stderr, "Out of memory in malloc!");
		exit(1);
	}
	retbuf[0] = (char) NULL;
	retbuflen = 0;
	buf[BUFSIZ] = (char) NULL;

	if (argc * sizeof(*argv) <= sizeof(env))
		argv = env;		/* use local buffer */
	else if ((argv = malloc(argc * sizeof(*argv))) ==
NULL) {
		fprintf(stderr, "Out of memory in malloc!");
		exit(1);
	}
	if ((argc = pread(fd, argv, argc * sizeof(*argv),
p->pr_argv)) > 0)
		argc /= sizeof(*argv);

	for (i = 0; i < argc; i++) {
		if (argv[i] == NULL || pread(fd, buf, BUFSIZ,
argv[i]) <= 0)
			continue;
		len = strlen(buf);

		/* If it's a bunch of spaces (zero'd out by
process), skip/trunc it*/
		sp = strchr(buf, ' ');
		if (sp == buf) 
			continue;
		else if (sp != (char *) NULL) {
			sp = buf + len;
			while (*--sp == ' ') {
				*sp = (char) NULL;
				len--;
			}
		}

		while (retbuflen + len + 1 >= argsize) {
			argsize *= 2;
			if ((retbuf = realloc(retbuf, argsize)) == NULL) {
				fprintf(stderr, "Out of memory in realloc!");
				exit(1);
			}
		}
		if (retbuflen != 0)
			retbuf[retbuflen++] = ' ';
		strcpy(retbuf + retbuflen, buf);
		retbuflen += len;
	}

	if (argv != env)		/* if we allocated argv */
		free(argv);			/* free it */

	return(retbuf);
}
	
/*
 * subproc to display environment variables
 */

showenv(int fd, struct prpsinfo *p)
{
	char *env[400];					/* hold environment */
	char buf[BUFSIZ];				/* temporary buffer */
	int i = 0;

	pread(fd, env, 400, p->pr_envp);
	while (env[i] != NULL && i < 400) {
		pread(fd, buf, BUFSIZ, env[i++]);
		printf("%s ", buf);
	}
	puts("");
}

/*
 * Main program
 *
 * fast - regular qps, and qps -d (to find stuck
processes waiting on 
I/O)
 *          qps -a (showargs), 
 *          qps -U (don't do NIS/NIS+ lookups),
 *          qps -s (show RSS, SIZE stuff)
 *        
 * medium - qps -e (has to munge through and grab out
environment 
stuff.
 *			uses read on open descriptor and has to do
pointers to pointers)
 *
 * slower - sorting of any kind. May get stuck if a
process is hung 
waiting
 *           for I/O. If so, run qps -d. (It's still
pretty darned 
fast)
 *          
 */

main(int argc, char *argv[])
{
	struct prpsinfo p;				/* process information
structure */
	struct passwd *pw;				/* hold password lookup
information */
	char rssinfo[80];				/* character resident set/memory
info */
	DIR *dirf;						/* directory file pointer */
	char nothing[2]=" ";			/* Just empty printing stuff
*/
	struct dirent *dirp;			/* directory pointer */
	int fd;							/* file descriptor */
	char c;							/* for getopt */
	int cnt, i;						/* counting and looping */
	int mapuid = TRUE;				/* Map userid to username? */
	register int min, maj;			/* Major and minor device
numbers */
	int (*func)(const void *, const void *) = NULL;		/*
sorting function 
pointer */
	struct qps pstruct[1024];		/* Process structure */
	char parg[15];					/* process name buffer */
	char *ppstr = "";				/* parent and pgid */
	char *pphdr = " PPID  PGID ";	/* PPID and PGID
headers */
	char *ppdelim = "";				/* delimiters for PPID and
PGID */
	char *ppdelimd="----- ----- ";	
	char ppbuf[40] = "";			/* hold actual ppid and pgid
*/

	/* flags */
	int debug = 0;					/* debugging flag */
	int showargs = 0;				/* Show arguments */
	int rss = 0;					/* Resident set size */
	int environ = 0;				/* show environment? */
	int showppid = 0;				/* show parent and pgid */
	int Allargs = 0;				/* show ALL arguments */

	while ((c = getopt(argc, argv, "AadertuUpPcmns")) !=
-1) {
		switch(c) {
			case 'A':
				if (rss) {
					puts("rss and argument showing are mutually
exclusive");
					exit(1);
				}
				Allargs = 1;
				break;
			case 'a':
				if (rss) {
					puts("rss and argument showing are mutually
exclusive");
					exit(1);
				}
				showargs = 1;
				break;
			case 'c':
				func = &sort_by_cpu;
				break;
			case 'd':
				debug = 1;
				break;
			case 'e':
				environ = 1;
				break;
			case 'm':
				func = &sort_by_mem;
				break;
			case 'n':
				func = &sort_by_pname;
				break;
			case 'p':
				func = &sort_by_pid;
				break;
			case 'P':
				showppid = 1;
				ppdelim = ppdelimd;
				ppstr = pphdr;
				break;
			case 'r':
				reverse = 1;
				break;
			case 's':
				if (showargs) {
					puts("rss and argument showing are mutually
exclusive");
					exit(1);
				}
				rss = 1;
				break;
			case 't':
				func = &sort_by_tty;
				break;
			case 'u':
				func = &sort_by_user;
				break;
			case 'U':
				mapuid = FALSE;
				break;
			default:
				usage();
				break;
		}
	}

	/* Set line buffering for output */
	setvbuf(stdout, NULL, _IOLBF, 1024);

	if ((dirf = opendir("/proc")) == NULL) {
		perror("couldn't open proc");
		exit(1);
	}
	(void) readdir(dirf); /* skip over . and .. */
	(void) readdir(dirf);
	
	/* Display headers */
	printf("%-7s %s%-3.3s %-8.8s %-4.4s %-4.4s %-17.17s",
"PID", ppstr, 
"TTY", "User", "%CPU", "%Mem", "Process Name");
	if (rss) /* print process size headers */
		printf(" %5s %5s %6s %6s\n", "SIZEP", "RSSP",
"SIZE_B", "RSS_B");
	else
		printf("\n");
		
	printf("%-7s %s%-3.3s %-8.8s %-4.4s %-4.4s %-17.17s",
"------", 
ppdelim, "---", "-------", "----", "----",
"-----------------");
	if (rss) /* print process size headers */
		printf(" %5s %5s %6s %6s\n", "-----", "-----",
"------", "------");
	else
		printf("\n");

	cnt = 0;
	/* open /procfs and scan through files one at a time
- each a process 
*/
	while ((dirp = readdir(dirf)) != NULL)  {
		sprintf(parg, "/proc/%s", dirp->d_name);
	
		if (debug)
			printf("process %s\n", dirp->d_name);

		if ((fd = open(parg, O_RDONLY)) < 0) {
			continue;
		}

		/* Grab process information/status */
		if (ioctl(fd, PIOCPSINFO, (void *) &p) < 0) {
			close(fd);
			continue;
		}
	
		/* map major and minor device numbers */
		min = minor(p.pr_lttydev);
		maj = major(p.pr_lttydev);

		/* Show process arguments with name */
		if (Allargs)  {
			pstruct[cnt].arglist = showallargs(fd, &p);
		} else if (showargs) {
			pstruct[cnt].arglist = strdup(p.pr_psargs);
		} else if (rss) {
			sprintf(rssinfo, " %5d %5d %6u %6u", p.pr_size,
p.pr_rssize,
				p.pr_bysize/1024, p.pr_byrssize/1024);
			pstruct[cnt].arglist = strdup(rssinfo);
		} else {
			pstruct[cnt].arglist = nothing;
		}
	
		/* Do we want to see parent and pgrp? */
		if (showppid)
			sprintf(ppbuf, "%5d %5d ", p.pr_ppid, p.pr_pgrp);

		/* Convert mem and CPU usage to percentage of
machine capacity */
		pstruct[cnt].pid = p.pr_pid;
		strcpy(pstruct[cnt].ppid, ppbuf);
		pstruct[cnt].mem = p.pr_pctmem * 100.0 / (float)
0x8000;
		pstruct[cnt].cpu = p.pr_pctcpu * 100.0 / (float)
0x8000;
		/* Get uid/uname */
		if (!mapuid || (pw = getpwuid(p.pr_uid)) == NULL) 
			sprintf(pstruct[cnt].user, "%-8d", p.pr_uid);
		else
			sprintf(pstruct[cnt].user, "%-8.8s", pw->pw_name);
		sprintf(pstruct[cnt].pname, "%-16.16s", p.pr_fname);
		/* check controlling tty if available */
		if (p.pr_ttydev == PRNODEV)
			sprintf(pstruct[cnt].tty, "--");
		else if (maj == 0)
			sprintf(pstruct[cnt].tty, "co");
		else
			sprintf(pstruct[cnt].tty, "%-.1d", min);
		if (func == NULL) {	/* no sorting - just do it */
			printf("%-7d %s%-3.2s %-8.8s %4.1f %4.1f %-16.16s
%s\n", 
pstruct[cnt].pid, pstruct[cnt].ppid, pstruct[cnt].tty,
pstruct[cnt].user, 
pstruct[cnt].cpu, pstruct[cnt].mem,
pstruct[cnt].pname, pstruct[cnt].arglist);
			if (environ)
				showenv(fd, &p);
		}
		cnt++;
		close(fd);
	}
	closedir(dirf);

	/* sorting applied */
	if (func != NULL) {
		qsort((void*) pstruct, cnt, sizeof(struct qps),
func);
	
		for (i=0; i < cnt; i++) 
			printf("%-7d %s%-3.2s %-8.8s %4.1f %4.1f %-16.16s
%s\n", 
pstruct[i].pid, pstruct[i].ppid, pstruct[i].tty,
pstruct[i].user, pstruct[i].cpu, 
pstruct[i].mem, pstruct[i].pname, pstruct[i].arglist
);
	}

	exit(0);
}


Please find below the Johan Hartzenberg discovery
-------------------------------------------------
This question prompted me to try and find the answer
myself, purely for
curiosity's sake.  I have discovered something very
very strange!

After scanning through a few man pages trying to look
for a system
function, I did a truss -aef -o /tmp/ps.truss.out ps
-o s -p $$

(using ksh, $$ gives 580 in the example below)

The way the execution seems to first check the
platform specific stuff 
and
then execute the correct "ps" command.

Then the ps started to open the files under
/proc/*/psinfo  ... Every
single one of them!  Only immediately after process
580 did it write a
single " S\n" to standard out.  See the extract from
the truss below.

My guess is that the odd getdents in between is to
cater for processes 
that
got created or terminated during the time ps had been
running.  My 
surprise
is at the seeming inefficient coding!  Surely when
specifying a list of
process IDs, ps only need to open and actually read
from the psinfo 
file
belonging to the relevant process(es)!  Granted the
/proc structure is 
in
memory, but does that justify going through 1000s of
processes and 
actually
making three system calls per process, just to print
the info for a 
single
process?  Back in programming class I learned that
calling a function 
is
CPU expensive, despite it's other advantages (Eg the
cost of stack
operations vs simplified coding).  To be sure I
compared the results
between Solaris 7 and 8, and both acts the same.  Also
I compared the
results between an E250 and an E10K domain, and both
were the same.

Any comments/thoughts?

> zactomud102 [/tmp] # more ps.truss.out
14693:  execve("/usr/bin/ps", 0xFFBEFD44, 0xFFBEFD5C) 
argc = 5
14693:   argv: ps -o s -p 580

[snip]

14693:  open("/proc/12657/psinfo", O_RDONLY)          
 = 4
14693:  read(4, "\f\0 B\b\0\0\004\0\0 1 q".., 416)    
 = 416
14693:  close(4)                                      
 = 0
14693:  open("/proc/608/psinfo", O_RDONLY)            
 = 4
14693:  read(4, "\b01 B\b\0\0\001\0\002 `".., 416)    
 = 416
14693:  close(4)                                      
 = 0
14693:  open("/proc/12862/psinfo", O_RDONLY)          
 = 4
14693:  read(4, "\f\0 B\b\0\0\004\0\0 2 >".., 416)    
 = 416
14693:  close(4)                                      
 = 0
14693:  open("/proc/12516/psinfo", O_RDONLY)          
 = 4
14693:  read(4, "\f\002\b\0\0\003\0\0 0E4".., 416)    
 = 416
14693:  close(4)                                      
 = 0
14693:  getdents(3, 0x10010F378, 1048)                
 = 1024
14693:  open("/proc/11552/psinfo", O_RDONLY)          
 = 4
14693:  read(4, "\f\002\b\0\0\003\0\0 -  ".., 416)    
 = 416
14693:  close(4)                                      
 = 0
14693:  open("/proc/13035/psinfo", O_RDONLY)          
 = 4
14693:  read(4, "\f\002\b\0\0\003\0\0 2EB".., 416)    
 = 416
14693:  close(4)                                      
 = 0
14693:  open("/proc/12581/psinfo", O_RDONLY)          
 = 4
14693:  read(4, "\b02 B\b\0\0\001\0\0 1 %".., 416)    
 = 416
14693:  close(4)                                      
 = 0
14693:  open("/proc/5937/psinfo", O_RDONLY)           
 = 4
14693:  read(4, "\f\0 B\b\0\0\005\0\017 1".., 416)    
 = 416
14693:  close(4)                                      
 = 0
14693:  open("/proc/5446/psinfo", O_RDONLY)           
 = 4
14693:  read(4, "\b\002\b\0\0\001\0\015 F".., 416)    
 = 416
14693:  close(4)                                      
 = 0
14693:  open("/proc/29660/psinfo", O_RDONLY)          
 = 4
14693:  read(4, "\b\0 B\b\0\0\001\0\0 sDC".., 416)    
 = 416
14693:  close(4)                                      
 = 0
14693:  open("/proc/14691/psinfo", O_RDONLY)          
 = 4
14693:  read(4, "\b\0 B\b\0\0\001\0\0 9 c".., 416)    
 = 416
14693:  close(4)                                      
 = 0
14693:  open("/proc/580/psinfo", O_RDONLY)            
 = 4
14693:  read(4, "\b\0 B\b\0\0\001\0\002 D".., 416)    
 = 416
14693:  close(4)                                      
 = 0
14693:  write(1, " S\n", 2)                           
 = 2
14693:  open("/proc/29964/psinfo", O_RDONLY)          
 = 4
14693:  read(4, "\b\0 B\b\0\0\001\0\0 u\f".., 416)    
 = 416
14693:  close(4)                                      
 = 0
14693:  open("/proc/14365/psinfo", O_RDONLY)          
 = 4
14693:  read(4, "\b\0 B\b\0\0\001\0\0 81D".., 416)    
 = 416
14693:  close(4)                                      
 = 0
14693:  open("/proc/7102/psinfo", O_RDONLY)           
 = 4
14693:  read(4, "\b\0 B\b\0\0\001\0\01BBE".., 416)    
 = 416
14693:  close(4)                                      
 = 0
14693:  open("/proc/12633/psinfo", O_RDONLY)          
 = 4
14693:  read(4, "\b02 B\b\0\0\001\0\0 1 Y".., 416)    
 = 416
14693:  close(4)                                      
 = 0
14693:  open("/proc/12609/psinfo", O_RDONLY)          
 = 4
14693:  read(4, "\f\002\b\0\0\003\0\0 1 A".., 416)    
 = 416
14693:  close(4)                                      
 = 0
14693:  open("/proc/12582/psinfo", O_RDONLY)          
 = 4
14693:  read(4, "\b0202\b\0\0\001\0\0 1 &".., 416)    
 = 416
14693:  close(4)                                      
 = 0
14693:  open("/proc/12610/psinfo", O_RDONLY)          
 = 4
14693:  read(4, "\f01 B\b\0\0\0\n\0\0 1 B".., 416)    
 = 416
14693:  close(4)                                      
 = 0
[snip]
Yahoo! - Official partner of 2002 FIFA World Cup
http://fifaworldcup.yahoo.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From s.q.eckhause at larc.nasa.gov  Sun May 26 23:14:48 2002
From: s.q.eckhause at larc.nasa.gov (Sarah Eckhause)
Date: Sun, 26 May 2002 23:14:48 -0400 (EDT)
Subject: SUMMARY: mounting and reading ISO CD image files
Message-ID: <Pine.GS4.4.10.10205262257350.14540-100000@express.larc.nasa.gov>

Solution:
First create a device using lof admin utility, lofiadm:
	# lofiadm -a /path/to/iso/file/image
lofiadm will return the device name
	/dev/lofi/1
Mount the iso image/device name just returned to your favorite mount point.
	# mount -F hsfs -o ro /dev/lofi/1  /mnt
Afterward, unmount the device and delete the iso device
	# umount /mnt
	# lofiadm -d /dev/lofi/1

Documentation:
% man lofiadm

Many thanks to everyone for their prompt replies, especially
William Hathaway   
Steve Bagdon
David Foster
John Martinez
Jonathan Andrews
Timothy Lorenc
Joel.Lee
Osama Ahmed         
Kailashnath Rampur 
Larry Snyder      
Scott Howard     
Simon-Bernard Drol
Vahid Moghaddasi 
David Cole       
-sarah
s.q.eckhause at larc.nasa.gov
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From jflanda at unavarra.es  Mon May 27 09:47:09 2002
From: jflanda at unavarra.es (Javier =?ISO-8859-1?Q?Fern=E1ndez?= Landa)
Date: 27 May 2002 15:47:09 +0200
Subject: SUMMARY: Undelete files in Solaris 8
Message-ID: <1022507229.15089.27.camel@s166m12>

Thank you all.

It's a bad thing doing an rm without a backup. Think twice before a rm
with wildcards. There's no such good thing as the undelete command of
msdos old days.

But you can try Coroner's Toolkit ( http://www.fish.com/tct ). It a
computer forensic set of tools. A command gets all available blocks in a
file system to a file and other command analyzes it. If you know how to
get you deleted files among all those blocks you're lucky.

Thanks again.

Javier Fernandez.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From dirk.boenning at efonds24.de  Mon May 27 10:23:31 2002
From: dirk.boenning at efonds24.de (Dirk =?ISO-8859-1?Q?B=F6nning?=)
Date: Mon, 27 May 2002 16:23:31 +0200
Subject: [SUMMARY] Routing of subnets
Message-ID: <3CF24163.1040701@efonds24.de>

Hello,

first at all thanx to all.

I forgot to mention that this system is working under Solaris 7. So the
syntax of route add 123.123.123.128/28 gateway_ip doesn't work.
Sorry for that (didn't thought that any differenses in this area of 
Solaris).

The file /etc/netmasks seems not to be visited by the route command, so 
the real winner is:

route add -net 123.123.123.128 -netmask 255.255.255.240 gateway_ip metric

This is tested and works on my system. To verify you could use
route get 123.123.123.128


Once again, thanx to all

Dirk.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From jonas.blaberg at cellnetwork.com  Mon May 27 11:22:02 2002
From: jonas.blaberg at cellnetwork.com (=?iso-8859-1?Q?Jonas_Bl=E5berg?=)
Date: Mon, 27 May 2002 17:22:02 +0200
Subject: SUMMARY: disksuite: where to put meta databases?
Message-ID: <E7491F05BCBA0B41B40E23AF1039FE1E88465A@exgbgmb01.europe.cellnetwork.com>

Hello!

I had to keep my promise to summarize this issue, even if it is a long time
ago...



> Where should the replicas for the state database ideally be put?
> Should the slice be put in the beginning, the middle or the end of a disk?
(for performance reasons and/or stabililty)

The metadbs aren't acessed alot, typically only when the machine
starts up and when changes are made, ie. disk failed, disk removed etc.

I don't believe that location on disk is an issue; the replicas are small and
as far as I know, they aren't accessed very often (read at boot and written
during changes which are infrequent).

replicas are small, and are only accessed during state changes or when
bringing things online. There should be no real performance problems
wherever you put it.

> Should all disks contain a slice for the state database?

Can, but in big systems, it can be an overhead..

Up to you. If you have 2 disks, yes. If you 100 disks, maybe not.

You'll want enough that you could lose some disks (or maybe a
controller) and keep at least 50%. With 3 controllers and lots of
disks, this is easy. With one controller and few disks, it's more
difficult. :-)

> How many replicas could be used without causing performance problems for
DiskSuite?

don't know, but I typically not going over a screen full of metadb output,
as I then have to scroll back :)

I'm not aware of any performance problems with disksuite having "too many"
replicas.

AFAIK, the main criteria with MetaDB is to have as much redundancy as
possible, but beyond that, I don't think there are significant "performance
issues" about where the things actually live.

I've never used DS with more than about 20 disks. Each contained a
replica and I didn't notice trouble.

Remember, the info in these db's basically never changes, so it's not like you
gather up a lot of overhead by having a few extra metadbs lying around.

> Should I dedicate a disk just to contain state database replicas for some
reason?

No, definately NOT.
1) wasting space,
2) no redundancy :(

You should not dedicate a disk to replicas; as well as creating a single point
of failure (lose the disk, lose the system), it's a waste of space. You can
easily fit 3 replicas in a 10MB partition.

No... what happens if you lose that disk?

That would be a bad idea. Loss of this single disk will cause pain.




> And finally some additional info from several guys:

You should try to balance the replicas across disks and controllers, but try
to imagine what would happen if a controller dies; i.e. if you have two
internal disks and 6 internal disks, don't put a replica on each disk as the
loss of the external disks would result in a loss of quorum.

Can't tell you about location on each disk. For the rest of your questions, it
would help to know how many disks you are going to use disksuite for. When I
have two disks I put three replicas on each, so that if one dies I still have
a quorum. You should put at least one replica on each disk for performance.
With three disks you might be able to get by using one or two per, but why not
use three?

The adviced way is to spread them over as many controllers as possible, after
that over as many disks as posible.
I typically "stage" all the disks in bigger setups with a 1 cylinder at
the start to prevent DBs etc. corupting the VTOC/partition table, and that's
where I put my metadbs.
If I have a few disks, I add more than one metadb per disk, and make sure
I have an odd amount of metadbs.


if you have 2 controllers and 8 drives total in a system, [drives 1-4 on C1,
drives 5-8 on C2] ensure to have at least one copy of metadb on devices under
each seperate controller, and likely prudent to have a few more copies than
that - say, a total of 4 metadb replicas, 2 per controller, one per drive, on
drives 1,3 and 6,8.

I dont know about best practise. My practise is
A small slice (minimal number of cynders) on each disk with each disk
containing at two copies of the data base in this slice. If there were many
disks involved, perhaps only one copy on each disk would be required.

A brief explanation about metadbs. You can place them into a small unused
partition, at the beginnig of any used partition ( if u have enough space,
of course), and even you can "steal" some space from swap; this is, you can
boot in single user mode, delete the swap partition using the command "swap
-d", repartition the disk using your new free space and construct the
partition you need, and finally boot the system and add swap space.
At least three copies of the metadb are needed. The explanation for this is
a little larger; I recommend you to read DiskSuite 4.2.1 administration
guide. This is available at docs.sun.com in pdf format.
Sorry for may grammar mistakes, I'm not very good at English.




many thanks to:

Hendrik Visage [hvisage at is.co.za]
Riddoch, John E SITI-ITDSEP3 [John.E.Riddoch at is.shell.com]
Eva Harlington [anaestuardo at hotmail.com]
Tim Chipman [chipman at ecopiabio.com]
Mike's List [mikelist at sky.net]
Darren Dunham [ddunham at taos.com]
Broun, Bevan [brounb at adi-limited.com]
eshafto at mac.com


/jonas


-----Ursprungligt meddelande-----
Fren: Jonas Bleberg
Skickat: den 16 april 2002 10:33
Till: sunmanagers at sunmanagers.org
Dmne: disksuite: where to put meta databases?


hello!
some questions about disksuite and how to organize things according to
security and performance.
Where should the replicas for the state database ideally be put?
Should the slice be put in the beginning, the middle or the end of a disk?
(for performance reasons and/or stabililty)
Should all disks contain a slice for the state database?
How many replicas could be used without causing performance problems for
DiskSuite?
Should I dedicate a disk just to contain state database replicas for some
reason?
etc.
Someone, somewhere has probably asked those questions before...
I will summarize. /jonas


Jonas Bleberg
  Cell Network Sverige AB
  Kruthusgatan 17,6
  S-411 04 Gvteborg
jonas.blaberg at cellnetwork.com
@office: +46-(0)31-707 69 85
not at office: +46-(0)709-95 00 68
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From makhlooq at batelco.com.bh  Tue May  7 04:12:59 2002
From: makhlooq at batelco.com.bh (Makhlooq)
Date: Tue, 7 May 2002 11:12:59 +0300
Subject: SUMMARY:Re: New H/D
Message-ID: <01aa01c1f59f$00580c20$8e70bcc1@DELL>

> Dear all,
>
> Thanks for all those who reply, I managed to find the problem. It was the
> modification of the slice, Labeling which was not saved after formating
the
> H/D.....
>
>
> Thanks again
>
> H. Makhlooq
>
>
> ----- Original Message -----
> From: "Makhlooq" <makhlooq at batelco.com.bh>
> To: <sunmanagers at sunmanagers.org>
> Sent: Tuesday, May 07, 2002 7:57 AM
> Subject: New H/D
>
>
> > Dear all,
> >
> > I have E220 Sun Machine with one H/D, I added another H/D to the
Machine.
> I
> > formated the H/D using format command, but when I try to mount it to a
> > directory I got this message :
> > mount: I/O error
> > try newfs I got :
> > /dev/rdsk/c0t8d0s0: I/O error
> > any one can help.
> >
> >
> > Hussain Makhlooq
> > STO System Management ( NI21 )
> > Office :+973-883063
> > Mobile : +973-9685646
> > Fax    :+973-9103063
> > _______________________________________________
> > sunmanagers mailing list
> > sunmanagers at sunmanagers.org
> > http://www.sunmanagers.org/mailman/listinfo/sunmanagers
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From kieran at esperi.demon.co.uk  Wed May 15 07:16:21 2002
From: kieran at esperi.demon.co.uk (Kieran)
Date: Wed, 15 May 2002 12:16:21 +0100 (BST)
Subject: SUMMARY Re: Messages flooding logs
In-Reply-To: <Pine.LNX.4.43.0205150416510.12910-100000@loki.wkstn.nix>
Message-ID: <Pine.LNX.4.43.0205151208481.12910-100000@loki.wkstn.nix>

On Wed, 15 May 2002, Kieran wrote:

> Hi,
>
> I have just been asked to start looking after a new box.
> First thing I see when I run dmesg is:
>
> May 15 02:04:58 box inetd[29061]: [ID 161378 daemon.error] exec/tcp:
> bind: Address already in use
> May 15 02:04:58 box inetd[29061]: [ID 161378 daemon.error] shell/tcp:
> bind: Address already in use
> May 15 02:14:58 box inetd[29061]: [ID 161378 daemon.error] exec/tcp:
> bind: Address already in use
> May 15 02:14:58 box inetd[29061]: [ID 161378 daemon.error] shell/tcp:
> bind: Address already in use
> (repeated many times, with 10 minute intervals).
>
> /var/adm/messages is similarly flooded.
>
> How do I stop the flood?  A quick look in /etc/inetd.conf on that box
> shows...
>
Thanks for the responses on this.

I received one suggestion to remove the tcp6 lines from /etc/inetd.conf,
and a number of suggestions to look for duplicate inetd processes,
either using ps or lsof.

ps did not show up a second copy of inetd, and lsof is not available on
this box, so I tried to restart with kill -HUP <pid>.

This actually failed (log message was:
May 15 11:42:39 epdb01 inetd[29061]: [ID 244026 daemon.notice] config:
100232/rpc/udp still active and was not reconfigured.
May 15 11:42:39 epdb01 inetd[29061]: [ID 268928 daemon.error]
rusersd/rpc/datagram_v,circuit_v: could not get transport information
May 15 11:42:39 epdb01 inetd[29061]: [ID 307641 daemon.error]
rusersd/rpc/datagram_v,circuit_v: tli_socket: No such file or directory
May 15 11:42:39 epdb01 inetd[29061]: [ID 161378 daemon.error]
printer/tcp: bind: Address already in use

Running /usr/sbin/inetd -s seems to have made the log messages
disappear.  Is this known behaviour, or do I file it under
"strange"?

(uname -a gives:
SunOS epdb01 5.8 Generic_108528-12 sun4u sparc SUNW,Ultra-Enterprise

Regards

Kieran
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From marco_greene at hotmail.com  Tue May  7 23:40:28 2002
From: marco_greene at hotmail.com (Marco Greene)
Date: Tue, 07 May 2002 23:40:28 -0400
Subject: SUMMARY A5200 Disk Array
Message-ID: <F225jpyr6gz1Lz7ccDo0000c272@hotmail.com>

Thanks to all who replied.Here is my original post:I have an E450 with a
A5200 disk array which has 10 disk drivesof 18GB each.I'm running Solaris
7 and Veritas Volume Manager 3.1.1.I want to add 4 disks of 36GB to the
A5200.Is this possible, will I run into any problem having both 18GB and
36GBdisks on the A5200?Also I noticed that the existing disks are not
located continuously in
theA5200 right now. Is there any rule to follow for the location of thedisks
in the A5200 array? Any web site or document where I can findthis
information? Here are the replies:
From: "Haywood, Steven" 
shaywood at hurricaneseye.comHiNot sure about the 36 gig thing (I'm likely
to find out though in a week or
two when my new 36's arrive) ;)As to the drive placement, yes there are
rules. The following bays NEED to
be occupied:
slots 0,5,10 in the front and 0,3,6,10 in the rear MUST have drives.This
comes direct from a Sun techie who inspected my kit :)Hope this helpsCheers
Steven================================From: Jay Lessert
It is possible, and should not cause any problems.  I'm doing this
with my A5200 right now (10x18GB and 12x36GB).> Also I noticed that the
existing disks are not located continuously in the
> A5200 right now. Is there any rule to follow for the location of the
> disks in the A5200 array?Hmmm, I don't know if there are official
recommendations or not.  I've
always tried to spread the heat load and weight out evenly, that seems
like a good idea.> Any web site or document where I can find
> this information?You can try http://sunsolve.sun.com/handbook_pub/.================================From:
mike at cluon.priv.at (Thomas 'Mike' Michlmayr) [...]
> Is this possible, will I run into any problem having both 18GB and 36GB
> disks on the A5200?i have this running, plus several 9 GB disks.> Also
I noticed that the existing disks are not located continuously in
> the A5200 right now. Is there any rule to follow for the location of
the
> disks in the A5200 array? Any web site or document where I can find
> this information?there actually is a document on sunsolve that
describes the minimum
installed disks (4 in the front, 5 in the back, in specific slots IIRC).
but, if your A5200 is working flawlessly, you most likely have a working
config. when i reconfigured ours (which only had 6 disks, all in the
front),
we ran into troubles really fast. if you have the required slots
occupied,
the rest can be placed in whatever order you prefer.
=====================================From: John Stoffel 
stoffel at lucent.comThere are no rules that I'm aware of, just stuff the
new disks in,
run:   drvconfig
   devlinks
   disks
   vxdiskadmAnd initialize the new disks.  I'd suggest that you put them
into
their own volume(s) for your sanity.  But it really depends on your
existing layout.John

------------------------------------------------------------------------

Send and receive Hotmail on your mobile device: Click Here
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From MahtoA at dime.com  Fri May 17 10:22:32 2002
From: MahtoA at dime.com (Amindra Mahto)
Date: Fri, 17 May 2002 10:22:32 -0400
Subject: SUMMARY:Re: UFS tuning
Message-ID: <sce4d9ec.065@dime.com>

 I thank following people for their suggestion in looking in several
directions which definitely gave some clue but the problem was solved by
something which was very unlikely. I changed the HBA as the last weapon and
guess what , it worked. Its absolutely fine now.

Thanks:
Mark Bergman
Buddy Lumpkin
Kieran
==================
Amindra K Mahto

Original Question is as below:


> > Hi Gurus,
> >
> > I have a strange problem in creation time of ufs file system in two box.
Both
> > the boxes are exactly identical in terms of Architecture, CPUs, Memory but
one
> > box takes less time than other. I have two boxes E-450, each has got 4
CPUs of
> > 400MHz and 2GB of RAM. Both box has got 8 (8.3GB) SCSI HDDs having same
speed
> > and access time. Now the problem is when I create a 2GB UFS file system,
one
> > takes 2 minutes whereas other one take more than 3 minutes. I am issuing
the
> > same newfs command in both boxes. Could you please help me in finding the
> > cause of this. Do I need to do any tuning. Just to add you that both boxes
are
> > having Solaris 2.8 with latest patches installed.
> >
> > Thanks!
> >
> > Amindra
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From ak94601 at yahoo.com  Fri May 17 15:02:53 2002
From: ak94601 at yahoo.com (Ath Kar)
Date: Fri, 17 May 2002 12:02:53 -0700 (PDT)
Subject: Summary ...Re: SunFire 880 ..SWAP
In-Reply-To: <200205162013.g4GKDB604344@star.starinc.com>
Message-ID: <20020517190253.54604.qmail@web14807.mail.yahoo.com>

Hi 
Thanks to Simon, Bertrand, Larye, Darren, Kanchan,
Todd, Mike, Lars...

It was too fast and Solution was quite simple one,
once I got the answer...

1. makefile 4 gb swaptmpfile
2. swap -a swaptmpfile
3. swap -d orig-swap-partition
4. change orig-swap partition with format command
5. swap -a new-swap-partition (Again add this swap)
6. swap -d swaptmpfile 
7. Change in /etc/vfstab file....

Lars had experienced the problem, but he then rebooted
server directly from Solaris 1/2. Then it was as
usual.

Thanks once again..

Ath
LAUNCH - Your Yahoo! Music Experience
http://launch.yahoo.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From inzel at tunku.uady.mx  Wed May 22 17:09:19 2002
From: inzel at tunku.uady.mx (Israel Novelo)
Date: Wed, 22 May 2002 16:09:19 -0500
Subject: Summary:RE: disk problems
In-Reply-To: <63C0F78C06E4D511B31E0002A55CA03C02D4AFA7@msg1.uslec.net>
Message-ID: <5.1.0.14.0.20020522160509.01e17e80@tunku.uady.mx>

Hi everybody, thanks for all who answer...

  The problem whit the disk is fixed... i made this:

1.- format > select disk
2.- partition
3.- modify
4.- free hog [0]
5.- and clean the past partition
6.- select every partition and assigned their sizes
7.- newfs -v /dev/rdsk/c0t8d0s0 we create a new file system
8.- mount  /dev/dsk/c0t8d0s0 /directory




At 16:29 22/05/2002 -0400, you wrote:
>You need to do the newfs on the raw device, i.e.
>
>newfs -v /dev/rdsk/c0t8d0s0
>
>Hope this helps,
>Rick Place
>Sr. Unix Systems Administrator
>US LEC of North Carolina
>
>
>
>-----Original Message-----
>From: Israel Novelo [mailto:inzel at tunku.uady.mx]
>Sent: Wednesday, May 22, 2002 4:20 PM
>To: sunmanagers at sunmanagers.org
>Subject: disk problems
>
>
>Hi dear partners:
>
>         I install a new disk in a E450 whit Solaris 8, and cant4n mount, i
>make
>this:
>
>   Format > format that disk
>
>newfs -v /dev/dsk/c0t8d0s0
>and send this I/O error
>
>later if i want mount that disk:
>
>mount /dev/dsk/c0t8d0s0 /directory
>
>and send I/O error
>
>and can4t mount...
>
>any ideas?
>
>thanks and i will summary...!!
>_______________________________________________
>sunmanagers mailing list
>sunmanagers at sunmanagers.org
>http://www.sunmanagers.org/mailman/listinfo/sunmanagers
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From ahoesch at smartsoft.de  Mon May 27 17:20:57 2002
From: ahoesch at smartsoft.de (=?ISO-8859-1?Q?Andreas_H=F6schler?=)
Date: Mon, 27 May 2002 23:20:57 +0200
Subject: Summary: Netra X1 as a router
In-Reply-To: <sh4rgtzfo4.fsf@lnxcu9.lns.cornell.edu>
Message-ID: <A2772924-71B7-11D6-BC08-003065CCA582@smartsoft.de>

Hi,

thanks to all that respondend to my problem and special thanks to 
Kamalan Govender who provided the last bit of the puzzle. The problem 
was caused by the second router in the 192.168.1.0 subnet. I had to add 
a dafault route to the DSL-Router on the net router and a route to the 
net router on the DSL-router. Then it worked.

Thanks,

   Andreas

Original message follows:
====================================================00
I am trying to setup a Netra X1 as a router without success so far. On 
the Netra I have the following files:

/etc/hostname.dmfe0:
=================
192.168.1.8

/etc/hostname.dmfe1:
=================
10.0.0.1

/etc/defaultrouter:
===============
192.168.1.254

/etc/resolv.conf:
=============
nameserver 192.168.1.1
nameserver 192.168.1.254

/etc/hosts:
========
127.0.0.1       localhost
192.168.1.8     neptun  loghost

The machine 192.168.1.254 is our gateway to the internet (router, DNS). 
On neptun (the Netra) I do

# nslookup www.google.de
Server:  merkur.smartsoft.de
Address:  192.168.1.1

Non-authoritative answer:
Name:    www.google.com
Address:  216.239.37.101
Aliases:  www.google.de

# ping www.google.de
ping: unknown host www.google.de

Why does nslookup succeed and ping fails? What could be wrong here?

When I do netstat -rn on the Netra I get

Routing Table: IPv4
   Destination           Gateway           Flags  Ref   Use   Interface
-------------------- -------------------- ----- ----- ------ ---------
192.168.1.0          192.168.1.8           U        1      1  dmfe0
10.0.0.0             10.0.0.1              U        1      1  dmfe1
224.0.0.0            192.168.1.8           U        1      0  dmfe0
default              192.168.1.254         UG       1      0
127.0.0.1            127.0.0.1             UH       2      6  lo0

This seems correct to me although I do not really understand all the 
details yet. I connected a machine with 10.0.0.2 to the second leg of 
the Netra. I can successfully do "ping 10.0.0.1" but "ping 192.168.1.1" 
fails. Shouldn't the Netra act as a router now? What am I missing here?

Thanks a lot for any hint!

Regards,

    Andreas
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From min at psde.mec.mei.co.jp  Tue May 28 04:13:28 2002
From: min at psde.mec.mei.co.jp (Min Oo Tint)
Date: Tue, 28 May 2002 09:13:28 +0100
Subject: SUMMARY: showmount: **** : RPC: Program not registered
Message-ID: <3C667B62E467F44DB51E01DC328A34311362DC@psde-server>

Hi,

only got one reply and many thanks to Pierre Zimmermann. 

it could be, that your system has some external file systems mounted via NFS
(the /etc/init.d/nfs.client was started)
but the RPC services (bind and portmapper) are not running anymore
(/etc/init.d/rpc)
The RPC services are basic services for NFS.
So you can start the RPC services manually: /etc/init.d/rpc start
then shutdown the nfs client: /etc/init.d/nfs.client stop
and stop the system: init 6

check the start and stop scripts in /etc/rc?/ may you have disabled or
removed some of them.

S* scripts for starting a service
K* for stopping 

they will be executed in alphabetical order e.g. K88 before K89.

---------------
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From doug_cooper at vizzavi.net  Tue May 28 05:52:18 2002
From: doug_cooper at vizzavi.net (doug_cooper)
Date: Tue, 28 May 2002 09:52:18 +0000 (GMT)
Subject: SUMMARY: Print error: Printer does not exist
Message-ID: <3cf35352.1109.16838@vizzavi.net>

Dear All,

Many thanks to Elizabeth Lee who suggested stopping &
restarting the lpsched daemon:

# lpshut
# /usr/lib/lp/lpsched

and Sean Burke, who suggested deleting and recreating the
print queue:

# lpadmin -x GEB50JR1
# lpadmin -p GEB50JR1 -s geb50j!GEB50JR1

I tried both of these on their own & in combination with
each other, but nothing has made any difference. If anyone
has any further suggestions, they are gratefully received,
but at the moment I'm putting it down to (not my) network or
h/w problems.

Thanks
Doug

Original post below:
----------------------------------------------------
Dear Sun-Managers,

I have an E5000 running Solaris 2.6 and have recently had
trouble printing to a remote printer. Attempts to send a
print job return message:

Error transferring print job 725
         check queue for (text at geb50j)

The printer is connected OK:

# ping geb50j
geb50j is alive

And lpstat reports correct information:

# lpstat -v GEB50JR1
system for GEB50JR1: geb50j (as printer text)
# lpstat -aGEB50JR1
GEB50JR1 accepting requests since May 21 14:28 2002

BUT lpstat -p gives this error:

# lpstat -p GEB50JR1
UX:lpstat: ERROR: Printer "GEB50JR1" does not exist.
          TO FIX: Use the "lpstat -p all" command to list
                  all known printers.

Attempts to enable the printer give this error:

# enable GEB50JR1
UX:enable: ERROR: Destination "GEB50JR1" is unknown to the
                  LP print service.

I am printing plain ascii text via lp. The printer has
worked fine for a long time, several months or even years,
but has just gone wrong yesterday. I can't find any
reference to these errors in the Sun documentation, so any
advice on how to fix this is much appreciated. I will
summarise.

Thanks
Doug
	
--------------------------------------------------------
Get it to go with Vizzavi Alerts - never miss a match,
soaps, news, reviews, celebrity comings and goings, daily
stars, weather and more.
http://www.vizzavi.co.uk
--------------------------------------------------------
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From irc2002 at hushmail.com  Tue May 28 08:41:19 2002
From: irc2002 at hushmail.com (irc2002 at hushmail.com)
Date: Tue, 28 May 2002 05:41:19 -0700
Subject: SUMMARY: DISK IDE 40GB and Solaris 7 
Message-ID: <200205281241.g4SCfJV06931@mailserver2.hushmail.com>

HI,

First of all, thanks sunners!

= Solaris 7 don't support disks over 32GB (only Solaris 8 +)
= You do not buy a IDE Maxtor disk . . .

My original question:

Hi ,

I have another problem.
Now I want to format a disk IDE Maxtor in an Ultra 10 with Solaris 7.
The format command: Cannot format this drive. Please use your Manufacturer supplied formatting utility.
I have already searched for a patch and I didinot find anyone.
Could anyone help me?

Thanks in advance

I.C.


Hush provide the worlds most secure, easy to use online applications - which solution is right for you?
HushMail Secure Email http://www.hushmail.com/
HushDrive Secure Online Storage http://www.hushmail.com/hushdrive/
Hush Business - security for your Business http://www.hush.com/
Hush Enterprise - Secure Solutions for your Enterprise http://www.hush.com/

Looking for a good deal on a domain name? http://www.hush.com/partners/offers.cgi?id=domainpeople
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From Alexander.Sarreiter at melo.de  Tue May 28 09:05:44 2002
From: Alexander.Sarreiter at melo.de (Alexander Sarreiter)
Date: Tue, 28 May 2002 15:05:44 +0200
Subject: [SUMMARY]  Problem installing a SUN 280R
Message-ID: <68D3780E25F21F40BE8B0717D9FA53BC19CF80@exmuc01.melo.de>

Hi everyone!

One suggestion was to boot from CD 1/2, rather than booting
from the Install CD. Since I have bootet from CD 1/2, I tried
to boot from the Install CD, it worked. ;-)

cu
alexs

> I have a problem installing a SUN 280R Server with Solaris 2/02.
> After the network configuration the popup window reads "Just a
> moment" and then can you wait forever, nothing happens.
>
> From the console:
>
> -------------------------snip-------------------------
> in.rdisc: No interface up
> Assertion failed: have_ext_net_info(rc)==True, \
> file main.c, line 931
> Abort - core dumped
> Starting remote procedure call (RPC) services: \
> sysidns done
> -------------------------snap-------------------------
>
> Seems that the sysid process hangs.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From walter at rotterdam.prutsclub.nl  Tue May 28 10:35:37 2002
From: walter at rotterdam.prutsclub.nl (Walter Heukels)
Date: Tue, 28 May 2002 16:35:37 +0200 (CEST)
Subject: SUMMARY: Ultra 5 system type in NVRAM?
Message-ID: <Pine.BSO.4.33.0205281634330.18787-100000@eithne.intranet>

And the winner is:

	80

Thanks to Matthew Stier.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From thesunlover2002 at yahoo.com  Tue May 28 10:36:05 2002
From: thesunlover2002 at yahoo.com (John Lee)
Date: Tue, 28 May 2002 07:36:05 -0700 (PDT)
Subject: SUMMARY: CPU extremely busy with high percentage of system
  processes
Message-ID: <20020528143605.74651.qmail@web14907.mail.yahoo.com>

Question:
I have a Sun system (an NFS client) having CPU problem
now. The system/kernel processes have consumed 50% to
70% of the CPUs. Here is the info from TOP: "CPU
states: 7.0% idle, 20.0% user, 72.9% kernel, 0.1%
iowait, 0.0% swap Memory: 1024M real, 479M free, 152M
swap in use, 1639M swap free". The user processes seem
fine in either 'top' or '/usr/ucb/ps -aux', and they
don't consume much CPU.

Solution:
No magic solution yet. I will post another question
about this issue.

Answers: Thank you all who answered.
james.brown at us.abb.com wrote:
Use snoop to check for NFS errors from other boxes.
That is a likely thing to look for. If you have a lot
of network traffic the kernel will consume the CPU
cycles.

"Kevin Buterbaugh" <Kevin.Buterbaugh at lifeway.com>
wrote:
If you're running Solaris 8, run prstat and see what
it says is using the most CPU. If you're running
Solaris 7 or earlier, you're going to have to rely on
top (although it's not supported by Sun and not always
100% accurate) for graphical info. How many CPU's does
the box have? If it has just one, context switching
may be your problem. Finally, does what "sar -u" and
vmstat tell you agree with what top says? If they
don't, believe sar and / or vmstat, not top.

"William Yodlowsky" <wyodlows at andromeda.rutgers.edu>
wrote:
We had a very similar issue here. We had moved a lot
of local disk data to a remote machine, and the
processing of the NFS and IP stack traffic alone was
consuming most of the CPU (same symptoms). Altering
the NFS read/write block size helped, but ultimately
the machine was just too slow to keep up, and the
hardware was upgraded.

"sajeev nv" <sajeev20 at rediffmail.com> wrote:
R u running web server or LDAP sort of applications ?
This can happen because of application or Oracle also.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

The following answer was from "Karl Vogel"
<vogelke at dnaco.net>. Thank you very much Karl !!

Answer:
Karl Vogel wrote:
We've run into this several times. Some suggestions
are below; our system is behaving much better.
First, I'd recommend upgrading to Solaris-8 if you're
not already there. It's stabler than 2.6 and has some
nice performance improvements. It comes with
filesystem journalling, which makes disk performance
absolutely *fly*. Solaris-8 also comes with a
different paging scheme, called priority paging in
earlier releases.
http://www.sun.com/sun-on-net/performance/priority_paging.html

-----------------------------------------------------------------------------
http://docs.sun.com/ab2/coll.709.2/SOLTUNEPARAMREF/
Overview of Solaris System Tuning
Tuning a Solaris System
Tuning the Solaris Kernel
Special Structures
Viewing System Configuration Information
kstats
Solaris Kernel Tunables
NFS Tunable Parameters
-----------------------------------------------------------------------------
You won't be able to get this unless you have a
SunSolve Online account.
http://sunsolve.sun.com/private-cgi/retrieve.pl?type=2&doc=stb/1442
White Papers/Tech Bulletins 1442
Delivering Performance on Sun: System Tuning
Greg Schmitz and Allan Esclamado, 30-Apr-1999
This document focuses on techniques for performance
tuning for the Sun computing environment.  It is aimed
at system administrators. Each chapter concentrates on
a different subsystem of the computing environment 
(e.g., Tuning the Solaris Kernel, Memory, Tuning Disk
Subsystems, etc.) and the specific things that can be
done to increase performance.
-------------------------------------------------------------------------
If you have an application that's an incredible swap
hog, or the system is really slowing down, try adding
the lines below to /etc/system and rebooting. I run
with these settings and they've never caused me
trouble.

*  Swap
*  System keeps 1/8th of all memory for swap, which is
too much for
*  a 4GB system. Reduce that to 32 Mbytes (4096 8K
pages).
set swapfs_minfree=4096

*  Memory management
* 
http://www.carumba.com/talk/random/tuning-solaris-checkpoint.txt
*  Tuning Solaris for FireWall-1
*  Rob Thomas robt at cymru.com, 14 Aug 2000
*
*  On firewalls, it is not at all uncommon to have
quite a bit of
*  physical memory.  However, as the amount of
physical memory is
*  increased, the amount of time the kernel spends
managing that
*  memory also increases.  During periods of high
load, this may
*  decrease throughput.
*
*  To decrease the amount of memory fsflush scans
during any scan
*  interval, we must modify the kernel variable
autoup.  The default
*  is 30.  For firewalls with 128MB of RAM or more,
increase this
*  value.  The end result is less time spent managing
buffers,
*  and more time spent servicing packets.
set autoup = 120

-------------------------------------------------------------------------
Check your most popular applications (using truss) for
the following:

* lots of kernel-level system calls, like open(),
read(), write(); all of these require an interrupt
plus kernel attention.
* lots of fork() or exec() calls to start new
processes; fork() under Solaris is extremely
expensive.
* lots of open files; a program called "lsof" can tell
you how many file descriptors are being used by
anything on your system.
* any opendir()/readdir() calls for walking through
directories to find a file; any given directory is
stored in a hash table, but the contents of the
directory have to be scanned linearly, so files in
large directories (~1000 or more files) will take much
longer to open or close.
* size of your inode caches, which keep track of
previously-accessed files.  Run the DNLC script below
as root to see your hit-rate percentage.  If it's
under 90-95%, you need to up the cache size. The
easiest way to do that is change maxusers in
/etc/system to a nice high number like 2048.

Run "mount" to see how your filesystems are set up. 
I'm pretty sure you can mount filesystems with
"noatime" turned on, meaning don't bother updating the
access time whenever a file is opened. We use this
under Solaris-8, and it makes a *huge* difference if
you're doing something to a large number of small or
medium sized files.

---------------------------------------------------------------------------
#!/bin/sh
#
# NAME:
#    dnlc
#
# SYNOPSIS:
#    dnlc
#
# DESCRIPTION:
#    "dnlc" reports on Directory name lookup cache
statistics from
#    the kernel.  This corrects a bug in vmstat.
#
#    To change the kernel values, add something like
this
#    to /etc/system and reboot.  Both "nnn" numbers
should be the
#    same.
#
#         set ncsize = nnn
#         set ufs_ninode = nnn
#
# AUTHOR:
#    Kimberley Brown - UKAC Kernel Support
#    comp.unix.solaris

PATH=/bin:/usr/bin:/usr/local/bin
export PATH

adb -k /dev/ksyms /dev/mem <<END
="**  Directory/Inode Cache Statistics  **"
="----------------------------------------"
ufs_ninode/D"Inode cache size"
ncsize/D"Directory name cache size"
ncstats/D"# of cache hits that we used"
+/D"# of misses"
+/D"# of enters done"
+/D"# of enters tried when already cached"
+/D"# of long names tried to enter"
+/D"# of long name tried to look up"
+/D"# of times LRU list was empty"
+/D"# of purges of cache"
*ncstats%1000>a
*(ncstats+4)%1000>b
*(ncstats+14)%1000>c
<a+<b+<c>n
<a*0t100%<n=D"Hit rate percentage"
="(See /usr/include/sys/dnlc.h for more information)"
END

exit 0
Yahoo! - Official partner of 2002 FIFA World Cup
http://fifaworldcup.yahoo.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From mikelist at sky.net  Tue May 28 11:57:08 2002
From: mikelist at sky.net (Mike's List)
Date: Tue, 28 May 2002 10:57:08 -0500 (CDT)
Subject: SUMMARY: showrev -p
In-Reply-To: <Pine.GSO.3.96.1020528100710.5568B-100000@apollo.sky.net>
Message-ID: <Pine.GSO.3.96.1020528105331.5568E-100000@apollo.sky.net>

Thanks to...
Dennis Peterson, jim.musso, Min  Oo Tint, John Malick, 
Dan Astoorian, Lars Hecking, and William Hathaway.

I needed SUNWadmfw (cd 1/2)...added the package, showrev now works.
some other useful command to see which library/ies missing.
ldd /usr/bin/showrev --or other commands you're having problem with.


- Mike

On Tue, 28 May 2002, Mike's List wrote:

> # showrev -p
> ld.so.1: showrev: fatal: libadmapm.so.2: open failed: No such file or
> directory
> Killed
> 
> So I read the FAQ #2 at sunfreeware.com and set my LD_LIBRARY_PATH
> # echo $LD_LIBRARY_PATH
> /usr/local/lib:/usr/openwin/lib:/usr/dt/lib
> 
> showrev -p stills doesn't work, what package/s or variable/s am I missing?
> 
> Running Solaris 8 02/02, core installation, with all the packages from
> FAQ #5 installed and recommended patch.  Checked the archive, and there's
> no core file in /var/sadm or /var/sadm/pkg.
> 
> 
> - Mike
> _______________________________________________
> sunmanagers mailing list
> sunmanagers at sunmanagers.org
> http://www.sunmanagers.org/mailman/listinfo/sunmanagers
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From nikul.raval.b at bayer.com  Tue May 28 10:44:41 2002
From: nikul.raval.b at bayer.com (Nikul Raval)
Date: Tue, 28 May 2002 10:44:41 -0400
Subject: SUMMARY: which Solaris 8 server to run Oracle8.1.7
Message-ID: <20020528143632.1C39420821@sunmgrs.cs.toronto.edu>

As always, thanks for the quick and helpful replies managers!

General consensus is to go for V880 over E450 for the task at hand.

I've pasted replies below.

Original post...
> We would like to replace our E3000 (EOL  system).  It will have Solaris 8 and
> will  basically be used to run Oracle 8.1.7 and access databases (about 22,000
> queries per day at the max usage).  There would be a Clarion EMC array
attached
> to it and will be accessed via fiber.
> It would also have to have faster I/O for queries and some room for future
> expansion.
> How do we  pick a server?  Right now we've E450 and E3500 as potential
> candidates.

Thanks Tim Chapman, Jay Lessert, Sydney Weinstein, Peter Ondruska,
Bertrand_Hutin,
Joon Martin Hanse

Replies in no particular order below...
------------------------0x0--------------------------------------------------

What about trying the new 880? It's relative cheap and has a good
internal bandwith (compared to 450). THE E3500 is a good machine, but I
don't think that U get the same bandwith from the SBUS MBA as from the
fast PCI card.

==============================================================================================

Have a look at the Fujitsu-Siemens PrimePower Series (Sparc/Solaris) :
http://www.ftsi.fujitsu.com/services/products/primepower/index.html
A model 600 should do.

==============================================================================================

Have a look at http://www.sun.com/blueprints/0302/suntone.pdf

==============================================================================================

in reality the E450 and the E4500 are also EOL'd.

The replacement for the E450 is the V880.  The replacement for the E4500
is the V4800.

Since the V880 costs less than the E450 it replaced, thats a simple
consideration.

We have clients running Oracle on V880's and are quite happy with it.

-- ----------------------------------------------------------------------


AFAIK, if you are pricing / purchasing hardware brand-new, you would be well
advised to avoid E3500 and get a V880 instead, since:

-price points are MUCH more competetive
-based on Ultra-3 CPU so less likely to be EOL'd "sooner"
-Ram, CPU upgrades are FAR cheaper. ie, a 8-way loaded V880 is ~ $100k USD list
vs
at least 2x for a loaded e3500, I believe, with 8 CPUs (slower - 450mhz instead
of
750!) and probably less ram as an added penalty
-it is possible that 8-way server will be overkill, I guess concurrency and
response time are the issues to consider, ie, how urgent is it that users get
responses "quickly" ; how many concurrent queries do you expect to have?
Probably
you can get a feel for this based on current database stats ....

It is also ?possible? that a quad-V880 will cost "not much more" than a
quad-e450
... and the scaling options are far more attractive.  Even a dual-CPU 880 might
be
adequate horsepower (1500mhz total, ie, dual 750) to start with .. and then you
still have room to add 6 more CPUs if required ...

of course, if you are repurposing existing hardware that your site owns already
-
(e450, 3500, etc) - then .. purchase models are all thrown out of whack :-)
Hope this helps a bit,

--------------------------------------------------------------------------
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From martin at schmitt.li  Tue May 28 12:24:15 2002
From: martin at schmitt.li (Martin Schmitt)
Date: Tue, 28 May 2002 18:24:15 +0200
Subject: SUMMARY: The evil LD_LIBRARY_PATH
In-Reply-To: <20020528143033.A497@schmitt.li>; from martin@schmitt.li on
  Tue, May 28, 2002 at 02:30:33PM +0200
References: <20020528143033.A497@schmitt.li>
Message-ID: <20020528182415.A1948@schmitt.li>

Okay guys,

first of all I must admit that I'm a little bit dumb. I failed to mention 
the OS I'm using. As many of you guessed correctly, my question was about 
Solaris 2.8. I even led some of you to believe the question was about Linux. 
Next time I'll try to be more precise.

The solution is:

	crle -l /usr/lib:/usr/local/lib
	crle -u

Since we're talking about an application that was compiled with GCC 2.95.3
I won't look into 64-bit stuff at this point.

The most speedy reply (took a single minute) was:

Da ich vom 21.05. bis 03.06.02 Urlaub habe, kann ich Ihre
E-Mail erst am 04.06.02 lesen.
Diese Mail wird nicht weitergeleitet.

Just kidding, this was an Out-Of-The-Office notification from Germany. ;-)

One comprehensive list from Dan Astoorian:

*snip*
In order from most- to least-preferred:

1) Compile your applications correctly.  The application should be told
at compile time, via the -R flag (or, if using gcc, "-Xlinker -rpath=...")
where to search for its shared libraries; it shouldn't have to rely on
an environment variable.

2) If you're using Solaris 8 or later (you didn't say in your message),
consult the documentation for the "crle" command.

3) In a pinch, you could link or copy the libraries in /usr/local/lib
to a directory in the standard library path, such as /usr/lib .

You're very correct that LD_LIBRARY_PATH should be avoided; I'll spare
you the laundry list of reasons.
*snip*

The laundry list is here:

http://www.visi.com/~barr/ldpath.html

Another useful link, the Sun compile FAQ:

http://inscoe.org/compilesun


A few suggestions pointed to LD_RUN_PATH e.g.:

*snip*
Rebuild 'rsync'.

This time, set 'LD_RUN_PATH' to '/usr/local/lib' prior to compiling. (Or modify
the Makefile to add a '-R/usr/local/lib' right after the '-L/usr/local/lib')

Either of the two methods above will encoded /usr/local/lib as an additional
library search path into the executable.

Note: This is a good practice for all executable you build, that depend upon
non-/usr/lib libraries.

Note: Sun has been doing this for /usr/openwin since Solaris 2.5, and CDE since
Solaris 2.6.
*snip*

Thanks to Peter Evans who got me started with a real-life crle command 
line copied and pasted from his japanese (aargh) terminal.

Thanks everyone,

-martin


* Martin Schmitt wrote/schrieb:

> Hi folks!
> 
> I'm suffering from LD_LIBRARY_PATH-itis. Fortunately it's only one
> directory that's contained therein: /usr/local/lib
> 
> I have added /usr/local/lib to LD_LIBRARY_PATH in /etc/profile, so on every
> interactive session the environment is set and everything works fine.
> 
> Unfortunately if it's a non-interactive session, things start to go awry.
> In my case, rsync is causing the trouble:
> 
> [martin at linuxbox martin]$ rsync --rsh="ssh" \
>                                 --rsync-path=/usr/local/bin/rsync \
>                                 solarisbox:/etc/hosts /dev/null
> 
> ld.so.1: /usr/local/bin/rsync: fatal: libpopt.so.0: open failed: No such
> file or directory
> Read from remote host solarisbox: Connection reset by peer
> rsync: connection unexpectedly closed (0 bytes read so far)
> rsync error: error in rsync protocol data stream (code 12) at io.c(150)
> 
> The reason is extremely clear: LD_LIBRARY_PATH is missing. One solution to
> get out of this would be to write some kind of wrapper script for rsync and
> put it into the --rsync-path option. Another one might be to create a
> statically linked rsync. I don't particularly like either option.
> 
> I recall reading (and not quite understanding) a text about why 
> LD_LIBRARY_PATH is bad and how it can be replaced. But as I said, I wasn't 
> able to grok it, and I can't recall where I saw it anyway. On my Linux boxes 
> I know /etc/ld.so.conf as the central place for adding library paths, so I'm 
> traditionally not too comfortable about LD_LIBRARY_PATH.  
> 
> How can I - in a nutshell - replace LD_LIBRARY_PATH by a constant and 
> system-wide setting?
> 
> Thanks in advance,
> 
> -martin
> 
> -- 
> There are 3 sides to every argument: Mine, yours, and the right side.
> _______________________________________________
> sunmanagers mailing list
> sunmanagers at sunmanagers.org
> http://www.sunmanagers.org/mailman/listinfo/sunmanagers

-- 
a man a plan a canal panama
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From TUrie at trueposition.com  Tue May 28 13:23:26 2002
From: TUrie at trueposition.com (Urie, Todd)
Date: Tue, 28 May 2002 13:23:26 -0400
Subject: SUMMARY: DiskSuite Mirror Failure Testing
Message-ID: <3166C01D5E20D411AE6900508BC76CE801FDB064@fstpkop1.trueposition.com>

Thanks to the following people:

Darren Moulding
Dave, bela at nivek.org
Tom Payerle
Pierre Zimmermann

The following suggestions were offered:

#---------------------------------------------------------------------------

The metaoffline command effectively fails the disk slice in question.  I am
not sure how convincing a test that is, of course.

Removing even hot swappable disks from a system before the system denotes
the
drives as "failed" can cause hardware damage, or so I have been told.
Having
done it a few times before learning of this, I believe it is one of those
one
time in a hundred or more type things, but worth bearing in mind.

When I wished to test my disksuite mirroring, I offlined all partitions then
pulled the disk, which was a reasonably good test in my opinion.  Not
offlining
the disks would be even better, but chances are the OS could tell a disk
failed
(and presumably would automatically offline it), and the (albeit small) risk
of HW damage was enough to drop that idea.

As for SNMP traps, I believe this is done in conjuntion with another
package.
I just use a cron job running metastat and looking for any suboptimal
metadevs.

Tom Payerle

#---------------------------------------------------------------------------

Hi Todd,
try cfgadm_scsi or cfgadm_pci to remove a disk from your system (see man
pages of cfgadm)
e.g.: cfgadm -x replace_drive c0::dsk/c0t0d0
or cfgadm -cf disconnect c0::dsk/c0t0d0

be sure that you have also created a boot record on the second disk, if the
first one is your boot disk!
e.g.:  installboot /usr/platform/`uname -i`/lib/fs/ufs/bootblk
/dev/rdsk/c1t0d0s0
and enter this as an alternative boot disk (second boot device) in the boot
PROM)

for logging we use syslog, disksuite writes its messages to syslog.

cu
Pierre

PS: there are already some usefull scripts: 
http://pixie.madstop.com/jumpstart/scripts/install-disksuite.fin

#---------------------------------------------------------------------------

> When I have two disks mirrored, is there anyway that I can simulate a
> failure of one of them, short of just pulling the cable on one of the
disks?

Nuke the VTOC on the mirrored disk ;) 

> Also, I thought that I read somewhere that DiskSuite will send SNMP traps
on
> failure, however, I can't seem to find where I read that.  Is this true?
If
> so, can someone point me to the docs?

yep. Disksuite can send snmp traps when shit happens. The mib is kinda 
limited in scope but it'll do the job.

http://docs.sun.com:80/ab2/coll.260.2/DISKSUITEUG/@Ab2PageView/idmatch(SERVI
CETASKS-30812)?DwebQuery=disksuite+OR+snmp&oqt=disksuite+snmp&Ab2Lang=C&Ab2E
nc=iso-8859-1#SERVICETASKS-30812

WOOO! UGLY URL! 


--

Dave.

bela at nivek.org

#---------------------------------------------------------------------------

Todd,

A little slow with this reply, but you never know you may not have got this
solution yet.

You could
fmthard -s /dev/null /dev/rdsk/cxtxdxs2

(this should blat the contents of the disk in question, failing the mirror)

Then use
metareplace -e dxx /dev/rdsk/cxtxdxs2 (your original mirror and disk)

Regards
Darren Moulding

PS I recommend you test on non live data 1st

Todd Urie
Unix Support Specialist
610-680-2231 Direct
turie at trueposition.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From wolf at uen.org  Tue May 28 15:16:38 2002
From: wolf at uen.org (Wolfgang Schwurack)
Date: Tue, 28 May 2002 13:16:38 -0600
Subject: SUMMARY: Logging File Systems
Message-ID: <3CF3D796.9F46AD7A@uen.org>

Thanks to all you that replyed

mount -o remount,logging /export
then add logging as an option in /etc/vfstab
man mount_ufs

--
Wolfgang Schwurack
Unix System Administrator
University of Utah/Utah Education Network
Tel: (801) 587-9444
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From mbialik at infinityhealthcare.com  Tue May 28 16:08:55 2002
From: mbialik at infinityhealthcare.com (Mark A. Bialik)
Date: Tue, 28 May 2002 15:08:55 -0500
Subject: SUMMARY: Off-line it, or Detach it?
Message-ID: <3CF3E3D7.8A74E931@infinityhealthcare.com>

Thanks to:

Guy Purcell, Scott Howard, Neil Harrison, James Brown, Dan Lorenzini,
Tom Payerle, Gregg Mackenzie, Richard Eisenman, and John Eisenschmidt.

Turns out the disk must have been bad.  Following the advice below,
I tried to metareplace the mirrors with themselves, but the resync
failed and I started getting SCSI errors.

So, I metadetached the mirrors on the problem disk, shutdown,
slapped in a new disk, partitioned the new disk with the same slice
info (FYI: I tried a fmthard with the info from the failed drive, but
since the new disk was a different type/geometry, this failed. So, I
recreated the partitions by hand... makeing sure they were slightly
larger than the old partitions).  Then, simply re-attaching the
mirrors rebuilt the info.

Thanks very much to everyone for their help.

Mark

=======================================================================
Sometimes, but it almost never tells you what the problem was, so you 
won't know how to fix it the next time it happens.

Personally, I'd try an approach somewhere in between 2 & 3 first.  If 
the disk has physical problems, then #2 is a waste of time.  But if the 
problems aren't severe enough to require replacement, then #3 is 
overkill--at least for now.  (If the problems are physical, I'd 
definitely want the disk replaced; it's just better to do replacements 
when you _want_ to than when you _have_ to.)

I'd metadetach the submirrors on the bad disk (all of 'em).  Then 
reformat the disk to find/remove bad regions.  And finally, metattach 
the submirrors again.  All of that can be done without taking any 
services down.  If format reports tons of bad blocks, or you continue to 
see SCSI errors, replace the disk.  You don't say what system houses the 
disk in question; if it's hot-swappable, you should be able to do a 
complete disk replacement & mirror resync while the system is up & 
running.

--
Guy
(guy at extragalactic.net)

=======================================================================
There's two real options you can take here...

1. Reattached the mirrors. The best way do do this is with metareplace 
:
        metareplace -e d2 c2t1d0s0
        metareplace -e d8 c2t1d0s3

2. Swap the disk.

Personally, I'd go for number 1 and see what happens.  If the disk
really is
bad, it will fail again either during the resync or shortly afterwards,
at
which point you'll be no worse off than you are now and you'll have to
take
options 2.

  Scott

=======================================================================
First thing to try with your disksuite problem would be to do a vurtual
replace of the dodgy metadevices

i.e.

for d1 submirror  do "metareplace -e d2 c2t1d0s0"
for d7 sunmirror do "metareplace -e d8 c2t1d0s3 "

A "metastat" command should show the mirrors syncing, 
there is no need to reboot....


Hope this helps

Neil Harrison

=======================================================================
Yes I have that happen all the time I don't know why. Just a slice of a
disk
will go off line but other sices are fine. This is how you correct it.

1) Bring the mirror into the main window

2) Right Click on the offeneding slice that is offline and click on
info.

3) Click on ENABLE

4) Commit transacction

If all is fine with the disk is should begin mirroring again and all
will be
fine. I hope this is actually your problem. In fact I just did it 5
minutes ago
myself.

<James Brown>

=======================================================================
The first thing I would try is to use format(1M) to "repair" the disk.
The safest way to do this is to use the "read" command of the "analyze"
menu.  If it finds a bad block it will attempt to "repair" it (actually
it maps it to a spare sector).  This is the default behavior unless you
change it using the "setup" command.  If the read pass goes through
without errors you might consider doing one of the write options.  In
this case you can use setup to limit the range of the test to the
affected partitions.  Since they are in "maintenance" mode, disksuite
will not be updating them while you run your test.

I have used this many times with good success.  However, sometimes it
does not work, so you must replace the disk.  If that is the case, you
need to use metadetach rather than metaoffline for all metadevices on
the affected drive, and then metareplace -e after the new drive is
installed and properly partitioned.

Regards,

Dan Lorenzini           Greenwich Capital Markets

=======================================================================
Don't believe this will work, but who knows.  Reminds me of the old joke
of what an IT person does when they get a flat tire--- turn off and
restart
the car to see if it goes away.

Assuming the disk is OK, I believe this will solve the problem.  You
could
also go a bit further; detach the mirror, then re-init the mirror and
reattach.
I would probably do the re-init since isn't much more work, and should
really
clean up any data corruption issues (assuming a good disk).  

There should not be any problem doing this even on root.  After all, the 
mirrors are bad, so should not be in use by anything anyway.  Even if
were
in use, this is the point of mirroring.

The question is whether the old disk is bad or not, and whether the cost
of a new disk exceeds the cost of a possible disk failure.  Since you
are
mirroring to begin with, sounds like an important system, and I would
tend to 
replace the disk (I might put the old disk to duty in a less critical
situation).

BTW, you should be able to offline the working mirrors on the problem
disk,
replace the disk (if not hot swappable, will require rebooting.  You
should
ensure that you have more than 50% of the database replicas on other
disks
before rebooting, and delete the replicas on the problem disk).  Then
run
metareplace for each of the mirrors and should start resyncing.

<Tom Payerle>

=======================================================================
I would be inclined to first try a fourth option:

 - metareplace the "bad" submirrors in place:

        metareplace -e d2 c2t1d0s0
        metareplace -e d8 c2t1d0s3

If disksuite kicks it/them back out again, you
probably do have something wrong with the disk,
but you could also try option #5:

 - detach/unmirror the bad submirrors (it's been
   awhile since I've had to try this, so I can't
   remember if it will let you detach a bad
   submirror...maybe with the -f option...I dunno);
 - metaclear the bad submirrors;
 - either fsck or newfs (your choice) the bad
   partitions, the idea being to "clean up" any
   residual filesystem bugginess;
 - metainit the bad submirrors;
 - metattach the new submirrors.

If that doesn't work, option #2 would be my next
choice, then option #3.

Option #1 doesn't work because the mddb retains its
state between reboots.  It would still think that
the components are bad.

Good luck.

Gregg Mackenzie

=======================================================================
I would probably do:

Detach on failing disk (metadetach -f ...)

Clear failing disk (metaclear ...)

Get rid of Replica Dbs on failing disk (metadb -d ....)

Edit /etc/vfstab and change back to standard device names

Shutdown, remove failing disk, reboot (and check that everything comes
up
OK)

Shutdown, put in a new disk (be sure its clean; if it happens to have
Replica Dbs on it from some other previous configuration you may have
some
trouble). Reboot.

Setup the mirror configuration again ...

Richard Eisenman

=======================================================================
This might be a little late, but I thought it might help.

We have some V880s running DiskSuite 4.2, and we've seen one quirk. When
we were building
the systems and we were rebooting a lot (screwing with kernel
parameters) we found if the
system came up and the mirrors were out of sync despite a notmal reboot,
they would be out
of sync every time we rebooted. So we'd reboot, DS would tell us they
need maintenance, we'd
metareplace the disk with itself, let it do a full rebuild until DS said
they were
consistant, then reboot again and the same disk would be out of sync. If
we detached the
mirror and reattached (letting it sync obviously) it would fix the
problem and every reboot
after that would come up clean. Strange, but I've seen it on a couple
different Solaris
installs on a couple different boxes.

Aside from that, DS is great. If you're still having problems it might
be worth detaching
and reattaching the mirror to see if it fixes the problem before you do
something crazy like
reboot.

Best,
John 

=======================================================================

Original Question:

Hello:

I have a problem with a DiskSuite 4.2 mirror, and I'd like some advice
on how to tackle the problem.  I have a few two-way mirrors.  I recently
discovered that some of the sub-mirrors went into a
"Maintenance/Critical" state.  One mirror is mounted as / and the other
/var

In each case, the failed sub-mirror is on the same disk.  However, the
same disk also has another submirror which is working just fine, so I'm
guessing the disk may not actually be bad (then again, it could become a
problem).

I have included my metastat, metadb, and syslog output detailing the
errors at the bottom of this email.  In each instance, the bad submirror
is on c2t1d0.  The Metadb I also had on this disk is bad, but I've got
six other ones spread across two other controllers.

My question is this:  What is my best approach?  I can see three
options:

1) Reboot and hope the problem clears itself up  :)  Does this sctually
work sometimes?

2) Offline the submirrors and then "online" them.  Since one of the
submirrors is for / I'm not exactly sure if this is a good idea.  If it
matters, the problem disk is not the primary boot disk.  Is this a good
option to try before breaking the root mirror and going through the
hassle?

3) Detach/Unmirror the root, reboot, edit the correct files, come up
unmirrored, slap in a new disk, etc.

Again, I'm not sure the disk is actually bad since another submirror is
OK.  But there could be some bad sectors.

This is my first problem under DiskSuite in about two years, so I guess
I;ve been pretty lucky.  It obviously saved my butt, and I don't want to
make matters worse by doing something stupid.  Any help is  greatly
appeciated.  I have an hour of scheduled downtime starting in about 8
hours :)

Will summarize.

Thanks very much,
Mark


# metadb -i
        flags           first blk       block count
     a m  p  luo        16              1034           
/dev/dsk/c0t0d0s7
     a    p  luo        16              1034           
/dev/dsk/c0t1d0s7
     a    p  luo        16              1034           
/dev/dsk/c0t2d0s7
     a    p  luo        16              1034           
/dev/dsk/c1t0d0s7
     a    p  luo        16              1034           
/dev/dsk/c1t1d0s7
     a    p  luo        16              1034           
/dev/dsk/c1t2d0s7
      W   p  l          16              1034           
/dev/dsk/c2t1d0s7

# metastat | more
d2: Mirror
    Submirror 0: d0
      State: Okay
    Submirror 1: d1
      State: Needs maintenance
    Pass: 1
    Read option: roundrobin (default)
    Write option: parallel (default)
    Size: 24578400 blocks
 
d0: Submirror of d2
    State: Okay
    Size: 24578400 blocks
    Stripe 0:
        Device              Start Block  Dbase State        Hot Spare
        c2t0d0s0                   0     No    Okay
 
 
d1: Submirror of d2
    State: Needs maintenance
    Invoke: metareplace d2 c2t1d0s0 <new device>
    Size: 35549760 blocks
    Stripe 0:
        Device              Start Block  Dbase State        Hot Spare
        c2t1d0s0                   0     No    Maintenance

d8: Mirror
    Submirror 0: d6
      State: Okay
    Submirror 1: d7
      State: Needs maintenance
    Pass: 1
    Read option: roundrobin (default)
    Write option: parallel (default)
    Size: 4097920 blocks
 
d6: Submirror of d8
    State: Okay
    Size: 4097920 blocks
    Stripe 0:
        Device              Start Block  Dbase State        Hot Spare
        c2t0d0s3                   0     No    Okay
 
 
d7: Submirror of d8
    State: Needs maintenance
    Invoke: metareplace d8 c2t1d0s3 <new device>
    Size: 4097920 blocks
    Stripe 0:
        Device              Start Block  Dbase State        Hot Spare
        c2t1d0s3                   0     No    Maintenance




May  9 08:57:22 emsdb3 scsi: [ID 107833 kern.warning] WARNING:
/pci at 4,2000/scsi at 1/sd at 1,0 (sd46):
May  9 08:57:22 emsdb3  SCSI transport failed: reason 'incomplete':
retrying command
May  9 08:58:27 emsdb3 scsi: [ID 365881 kern.info] /pci at 4,2000/scsi at 1
(glm3):
May  9 08:58:27 emsdb3  Cmd (0x708fc320) dump for Target 1 Lun 0:

May  9 08:58:51 emsdb3 scsi: [ID 107833 kern.warning] WARNING:
/pci at 4,2000/scsi at 1/sd at 1,0 (sd46):
May  9 08:58:51 emsdb3  Error for Command: write(10)               Error
Level: Fatal
May  9 08:58:51 emsdb3 scsi: [ID 107833 kern.notice]    Requested Block:
12028560                  Error Block: 12028560
May  9 08:58:51 emsdb3 scsi: [ID 107833 kern.notice]    Vendor:
SEAGATE                            Serial Number: 3AK0E8CY
May  9 08:58:51 emsdb3 scsi: [ID 107833 kern.notice]    Sense Key: Not
Ready
May  9 08:58:51 emsdb3 scsi: [ID 107833 kern.notice]    ASC: 0x4
(<vendor unique code 0x4>), ASCQ: 0x1, FRU: 0x2
May  9 08:58:51 emsdb3 md_stripe: [ID 641072 kern.warning] WARNING: md:
d1: write error on /dev/dsk/c2t1d0s0
May  9 08:58:56 emsdb3 md_mirror: [ID 104909 kern.warning] WARNING: md:
d7: /dev/dsk/c2t1d0s3 needs maintenance
May  9 08:58:56 emsdb3 md_mirror: [ID 104909 kern.warning] WARNING: md:
d1: /dev/dsk/c2t1d0s0 needs maintenance
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From mira743 at go.com  Tue May 28 16:43:18 2002
From: mira743 at go.com (Mira)
Date: Tue, 28 May 2002 13:43:18 -0700 (PDT)
Subject: [SUMMARY] Timeout waiting for ARP/RARP packet
Message-ID: <5095434.1022618598874.JavaMail.mira743@gomailjtp02>

Dear SUN Managers : 
 
 I really appreciate all the responses I got from so many great people on this list. Most of you got it right. The problem was that the "diag-device" was set to "net". When I changed it to "disk" (setenv diag-device = disk), the timeout problem got solved. Also, I set diag-switch? to "false" as pointed out by David Foster to be the right solution (to keep the system off the diagnostics mode).
 
I received more than 30 responses. Below are some of the people who answered first (I really do appreciate also others who are not mentioned below) : 
 
 
 Alex Shepard 
 Darren Dunham 
 Sean Berry 
 Graham Wood 
 Valery Glinski 
 Ray Brownrigg 
 David Foster 
 Amit Mahajan 
 John England 
 Tony Walsh 
 
 Thanks, 
 Mira 
 
 
 Original Post : 
 
 Dear Sun Managers, 
 
 I was installing Solaris 2.5.1 11/97 (after using the Operating Environment CD) on a 440 MHz U10 and after rebooting the workstation, the following message 
appears always : 
 
 timeout waiting for ARP/RARP packet 
 
 I could see that it was trying to boot the U10 off the net , so I tried to force it to boot from disk with the OBP commands : 
 
 setenv boot-device disk 
 boot disk 
 
 but still got the same result after rebooting. 
 
 Please advise as to what could be the problem. 
 
 Thanks, 
 Mira. 

___________________________________________________
GO.com Mail                                    
Get Your Free, Private E-mail at http://mail.go.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From gene.g.beaird at mail.sprint.com  Tue May 28 17:40:24 2002
From: gene.g.beaird at mail.sprint.com (gene.g.beaird at mail.sprint.com)
Date: Tue, 28 May 2002 16:40:24 -0500
Subject: Partial SUMMARY: kernel warning message on a Netra T1
Message-ID: <H000199e1208bb51.1022622024.daopmp01@MHS>

Many months ago (almost a year ago, to be exact), I posted the 
below-noted query to this list regarding a problem we were experiencing 
on our Netra T1s.  I got several replies, most of which were like "We 
have the same problem, please let us know when you get a solution."

The reason I have not posted a SUMMARY to the Sun Managers list is that 
I don't have a specific solution.  I have found three possibilities, 
two of which were supplied by others on this list.  
_____________________
One person, (thanks Doug Martin-Vegue) had this quick solution;  

What he did was comment out the "lomp lomp" line in /etc/driver_aliases 
and the error on his system went away.  

That may not have cured the problem, but it stopped the messages on his 
system with no apparent side affects.  

_____________________
Kirk Gilliam ran the default upgrade via a tip command through a 
network cable attached to the LOM A port on the Netra.  I tried this, 
but it turned out that we had a bad cable (bad cable, no donut!!).  
Here is what Kirk did;

This is how I got the "lom" to work not using a dumb terminal, but on 
the back of another Sun:

        1. Attached the "Silver Null 25 pin with RJ-45 interface on 
back" to a "ttya or ttyb" 		port on the back of a 
        Sun 220r or 420r. 
        2. I attached a standard "cat 5" cable from that interface on 
the Sun 220r (420r) to 		the "lom A port" on the    Netra T1.

        3. From the Sun 220r (420r) I opened a terminal window and type 
the following: "$ tip 		/dev/ttyb -9600". 
                a. It then says "connected". 
                b. I then hit the <enter> key and I get "hostname 
console login" 
        4. I login from there. Here is a cut and paste of how I am 
connected: 

------------------------------------------------------------------------
--- 
        carl:/export/home/c6941 $ tip /dev/ttyb -9600 
        connected 

        ftp1 console login: some_userid 
        Password: 
        Last login: Fri Jan 11 10:57:47 from xxxxxx.corp.bc 
        This is a private system, unauthorized use is prohibited. 
        ftp1:/home/c6941 $ 
------------------------------------------------------------------------
--- 

The way I upgraded the lom with the patch was through a normal telnet 
session to the netra. But when I ran the "lom -G default" command and 
it ran through it's upgrade, I was in through the "lom console" using 
the method above.
__________________

What we did on one system, but haven't had a chance to back up and 
verify the process is this;

Install/re-install the Lights Out Management 2 software off of the 
Solaris 8 Supplement CD, then patch the machine with the LOM patch 
110208-09 (or better, 110208-15 is current).  That made the error 
messages go away on the machine we patched.  As I said, we haven't had 
a chance to test it on another system yet (other people's priorities, 
you know).  

Hope this helps.  Sorry for the delay, but thanks for the help along 
the way.

Gene Beaird 

_________________________
Original message follows:

> 
> I have just installed the 8_Recommended on one of our Netra T1 
servers.
>  Looking at the /var/adm/messages after reboot, I find the following
> line;
> 
> [ID 723599 kern.warning] WARNING: Driver alias "lomp" conflicts with 
an
> existing driver name or alias.
> 
> Checking SunSolve for this, I found a patch 110208-09 (SUNWlomm) was
> recommended and installed it also.  I still get the same kernel
> warning.  The machine seems to be running okay, but it would sure be
> nice to fix this message.  Has anyone run across this message before?
> Did you fix it?  If so, how?  Did you live with it?  Thank you for any
> info.
> 
> Gene Beaird
> Systems Integrator V
> Sprint E|Solutions ISS
> Houston, Texas
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From markhj01 at bahamut.thelair.dynamic-site.net  Tue May 28 18:05:42 2002
From: markhj01 at bahamut.thelair.dynamic-site.net (Dragon King)
Date: Tue, 28 May 2002 18:05:42 -0400 (EDT)
Subject: SUMMARY: Watchdog Reset
In-Reply-To: <EC719BFDBB12D4119D8F00062939D3E30B43B459@ntd-ksaver2.ksaver.pbz.hr>
Message-ID: <Pine.GSO.4.10.10205281745560.6357-100000@bahamut>

Thanks to the 2 people who responded on a holiday weekend:
Ric Anderson
Damir Delija

They suspected either battery over 2 years old and starting to fail or a
powersurge that passed thru the backup.. After testing the battery (unplug
as well as test switch) I'm calling this a fluke and hoping it doesn't
happen again.
Thanks again,
Jeff

--

This message brought to you courtesy of the Dragon King.

http://thelair.dynamic-site.net
dragonking at carolina.rr.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From ed.rolison at itc.alstom.com  Wed May 29 03:41:15 2002
From: ed.rolison at itc.alstom.com (ed.rolison at itc.alstom.com)
Date: Wed, 29 May 2002 08:41:15 +0100
Subject: SUMMARY: Automounter oddities.
Message-ID: <OFFA6092DE.EA706BCD-ON80256BC8.002A3AB5@transport.alstom.com>

Thanks to Charles Homan and Pierre Zimmerman.

The solution, was of course, to alter the automount maps so /SiteWide/home
was mapped
as a whole directory, rather than mapping user directories individually.
Looking at the config
we had (that I transplanted and then did a search and replace) it should
never have worked...
So for the moment, I'm calling it gremlins :)

Cheers.
Ed Rolison



CONFIDENTIALITY:
This e-mail and any attachments are confidential and may be privileged. If
you are not a named recipient, please notify the sender immediately and do
not disclose the contents to another person, use it for any purpose, or
store or copy the information in any medium.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From Siebe_Jamnagar at ril.com  Wed May 29 04:50:03 2002
From: Siebe_Jamnagar at ril.com (Siebe_Jamnagar at ril.com)
Date: Wed, 29 May 2002 14:20:03 +0530
Subject: SUMMARY: CDE - mounting partition
Message-ID: <OF6253FA31.58007148-ON65256BC8.00301507@ril.com>

Hi,

I could resolve the problem myself.  ( Thanks to the 'ldd' command )

I have checked up the dynamic dependencies using

#ldd /usr/dt/bin/dtlogin

This has shown usage of a library file from my slice7. The LD_LIBRARY_PATH
was corrected accordingly and the problem vanished.

Thanks ...


Original message
==========
Message: 15
Subject: CDE - mounting partition.
To: sunmanagers at sunmanagers.org
From: Siebe_Jamnagar at ril.com
Date: Tue, 28 May 2002 17:49:10 +0530

Hi gurus,

My machine is Sun E250, Solaris 2.6

Currently when CDE starts up, it make use of slice7 of the local disk (
target 0 ).  My problem is Im using this slice exclusively for some
specific tasks, which needs mounting and unmounting. While unmounting
slice7, it shows slice is busy and "fuser -c" reveals dtscreen program is
utilizing this slice.

Is it possible to change this. Like forcing it to make use of any other
slices.

Thanks in advance,

JS
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From jrodriguez at intellinet-tech.com  Wed May 29 05:12:46 2002
From: jrodriguez at intellinet-tech.com (Jeremy Rodriguez)
Date: Wed, 29 May 2002 05:12:46 -0400
Subject: SUMMARY: A1000 RAID
Message-ID: <HMEIJMEKLHGNJPCDCEBOMEFHDBAA.jrodriguez@intellinet-tech.com>

First,
A Big Thank You to all who responded: Patricia Mora, BB, Peter Evans,
Alexander Rogov, Dennis Martens, Thamara Wanigatunga
Question?
I am trying to reinstall an OS (Solaris 5.7) on a A1000. I have just added
> 4-18g drives for a total of 8. The problem is that my A1000 is seeing -
> c0t0d0 at install and not c0t0d1 or seeing them both as one drive. I
believe
> it should see 8-18g drives. It just recognizes the original 4-18g. Note
> these drives are from two different vendors IBM and Seagate, both used on
> most sun boxes.

Summary:
My main goal was to create one large LUN. I had already backed the file
systems so losing data was expected as well as a fresh install of Solaris. I
had to install the Raid Manager software and scsi card onto another sun box.
Bring up the gui, blow away the LUN's (which is why I had to switch boxes),
save, then hook it back up to the original machine and reinstall the OS as I
was planning.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From sunmanager at nc.rr.com  Wed May 29 08:27:31 2002
From: sunmanager at nc.rr.com (Vic Engle)
Date: Wed, 29 May 2002 08:27:31 -0400
Subject: SUMMARY: network traffic
Message-ID: <3CF4C933.5020502@nc.rr.com>

Thanks for all the excellent answers! Most people recommended ntop. Some 
suggested kstat for solaris 8 and netstat -k interface. Ntop in 
interactive mode looks to be a pretty good tool.

My original post:

Hello All,

Is there a command that allows real time monitoring of network traffic 
through a specific interface? I need to measure bytes or kbytes per 
second and netstat seems to report only in packets which can vary 
significantly in size.

Thanks,

Vic Engle
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From vk_sol at yahoo.com  Wed May 29 09:11:26 2002
From: vk_sol at yahoo.com (vijay)
Date: Wed, 29 May 2002 06:11:26 -0700 (PDT)
Subject: SUMMARY : Sudoers error(not allowd to execute the command)
Message-ID: <20020529131126.58456.qmail@web10602.mail.yahoo.com>

Hi All, 

Thanks to veryone who replied. 
Pierre 
Jason 
Matthew 


I have two scenario : 

1) Two uers x and z. give the su permission to x for z user only. So that x can su only to z. 

Add the following line in sudoers 

X ALL = (root) /usr/bin/su  z 

OR 

Host_Alias MYSVR = myserver 
Cmnd_Alias SU = /usr/bin/su - z 
X MYSVR = (root) SU 

x must use - with the command su. see the few lines of a mail which I received from Pierre 

>su without - is useless in your case. 
>It would be the same as: sudo -s z 
>(which starts a shell with z1 ownership) su - z starts a shell and executes all the profiles (setting >the environment vars) of z but if you really want to have su without - , then you have to define it in sudoers: 

> X ALL = (root) /bin/su - z, /bin/su z

> (root) is valid for both commands. Note that sudo >will only grant permissions for commands, which are  >exactly match the definition in your sudoers file. 



Now If you want user x should execute any the command as z then add following line 

Defaults:x set_home,runas_default=z 
x ALL = (z) ALL 

x should then call: sudo or sudo s 
x can run ALL progs as z on ALL systems. (ie: sudo u z /bin/ls ) 
This has the advantage of logging all actions. 
Substituting the '-s' option give a shell as that user. (ie: sudo -u z -s ) 
Adding the '-H' option also sets HOME to z's home directory. (ie: sudo -u z -s -H) 


2) If there are a two groups x and y. In x group i have three users u1, u2, u3 In y group i have five users z1, z2, z3, z4, z5. Now in this case any user of x gruoup shoud be able su any member of group y. 

Add following lines in sudoers file. 

# User aliases 
User_Alias X = u1,u2,u3 

# Runas alias specification 
User_Alias y = z1, z2, z3, z4, z5 

# Host aliases 
Host_Alias MYSVR = myserver 

# Command aliases 
Cmnd_Alias SU=/usr/bin/su - z1, /usr/bin/su - z2, /usr/bin/su - z3,.... 
OR 
Cmnd_Alias SU=/usr/bin/su  z* 

# Override builtin defaults 
Defaults:X set_home,runas_default=y
OR
Defaults:X !aunthenticate 

# User specifications 
X MYSVR=(root) SU 


Once again Thanks for giving your time. 
Regards, 
Vijay 



Original Post: 

I am using Solaris 2.6. Installed Sudoers version 1.6.3p7. I have two users x and y. Now I want to give su permission to x user for only y user. So x should su only to y. 

So I added following line in suoders 

x ALL = /usr/bin/su y 

But that didn't worked. Still it gives me following error 
Sorry x is not allowed to execute "/usr/bin/su y " as root on Server. 
Yahoo! - Official partner of 2002 FIFA World Cup
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From tim at load.com  Wed May 29 15:50:15 2002
From: tim at load.com (Timothy Lorenc)
Date: Wed, 29 May 2002 14:50:15 -0500
Subject: SUMMARY: SUN and Oracle performance design question
Message-ID: <000001c2074a$0e7d6b30$250a10ac@tp770x>

[Sorry if this is a duplicate, I did not see this message go through the
first time I sent it.]

I would like to thank the following individuals for responding to my query:

Luis Aguilar [laguilar at transformpharma.com]
topher [topher at findtopher.com]
Vberg Mats [mats.oberg at tietoenator.com]
Jeff Kennedy [jlkennedy at amcc.com]
Tim Chipman [chipman at ecopiabio.com]
Tristan Ball [tristanb at vsl.com.au]


From RABENOJA at prodigy.net  Wed May 29 22:52:15 2002
From: RABENOJA at prodigy.net (RABENOJA at prodigy.net)
Date: Wed, 29 May 2002 21:52:15 -0500
Subject: Summary - Expect Script
Message-ID: <AA-415EBA3E1B65493D5297DDC3D7710F60-ZZ@homebase1.prodigy.net>

> I want to thank all those who responded. I had an 
error in my script
> below (expect -r "$prompt" it should be 
expect "machinename%" - sorry
> about that). Josh Glover's hint's helped fix the 
problem.
> 
> 
> Have you tried adding a short sleep before sending 
the password? This 
> has helped me before.
> 
> e.g.:
> 
> > #!/usr/local/bin/expect
> > log_file -noappend "telnetoutput.txt"
> > spawn telnet 138.10.12.144
> > expect "ogin: "
> > sleep 1; send "dude\r"
> > expect "assword: "
> > sleep1; send "dude\r"
> > expect -r "$prompt"
> > send "cat /var/adm/messages | grep keyword"
> > send "exit\r"
> 
> Worth a shot, maybe.
> 
> 
> -- 
> Josh Glover <jmglov at incogen.com>
> 
> Associate Systems Administrator
> INCOGEN, Inc.
> 
> 
> 
> 
> Thank you,
> Bob Abenoja
> PBIS Tech Supoort - San Ramon CA
> 925.806.4782
> 
> 
> Original Message:
> ################
> 
> Dear Sun Gurus,
> 
> I have just installed expect 5.25 along with tcl-8.0 
> and tk-8.0 on a solaris 2.6 sparc 5.
> 
> I modified a script I saw posted over a month ago. 
> Here is a simple copy of my script:
> 
> #!/usr/local/bin/expect
> log_file -noappend "telnetoutput.txt"
> spawn telnet 138.10.12.144
> expect "ogin: "
> send "dude\r"
> expect "assword: "
> send "dude\r"
> expect -r "$prompt"
> send "cat /var/adm/messages | grep keyword"
> send "exit\r"
> 
> 
> The script gets to the password part and stops. The 
> prompt is then returned, almost looking like it 
tried 
> to insert the prompt for the password. I know the 
> password is correct and I also tried another 
account. 
> Can an Expect script expert help me figure this out.
> 
> Thank you,
> Robert Abenoja
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From sdevine at msu.edu  Thu May 30 08:12:00 2002
From: sdevine at msu.edu (Steve Devine)
Date: Thu, 30 May 2002 08:12:00 -0400
Subject: SUMMARY: Question about redirect
Message-ID: <02053008120000.24943@jax.cl.msu.edu>

Thanks to all ... your responses were greatly appreciated.
"Ya wouldn't  'a thought this would be so hard.."

Original problem:
When using scp with a script I was unable to capture its output.
scp normally returns a 'progress meter' style of message .
This doesn't seem to be able to be redirected in the 
usual manner. 
This would not work:
/usr/local/bin/scp -p /dir/dir/file.tar root at server:/dir/dir/  >/tmp/ouptut
Nor did about 5 other styles of using redirect. 
So at the advice of those acknowledged below I had scp return 
a exit status and crafted a email to be sent accordingly.  

---------------------------------------
#!/bin/sh
/usr/local/bin/scp -p /dir/dir/file.tar root at server:/dir/dir/  
hold=$?
if [ $hold = 0 ];
then
echo "Files transferred" >/tmp/output
/bin/mailx -s "File Transfer "  me at msu.edu </tmp/output
else
echo "File transfer has failed" >/tmp/output
/bin/mailx -s "File Transfer Failed"  me at msu.edu </tmp/output
fi
------------------------

David Morris <davey at arklight.demon.co.uk>
system administration account <sysadmin at astro.su.se>
Darren Dunham <ddunham at taos.com>

Thanks to All

-- 
Steve Devine   
Michigan State University



To the optimist, the glass is half full.
To the pessimist, the glass is half empty.
To the engineer, the glass is twice as big as it needs to be.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From cg98ah at yahoo.ca  Thu May 30 09:47:32 2002
From: cg98ah at yahoo.ca (Chunhang Gong)
Date: Thu, 30 May 2002 09:47:32 -0400 (EDT)
Subject: Summary: how to copy filesystems from an old disk to a new one
  using ufsdump
Message-ID: <20020530134732.27682.qmail@web10603.mail.yahoo.com>

Hello to all,

Many many thanks to all you for your time and your
help.

The solution:

method 1, backup the whole filesystems structure from
an old disk to the new one:
  
example: (new disk is c0t1d0, old disk is c0t3d0)
 
 # mount /dev/dsk/c0t1d0s5  /backup
 # ufsdump 0f - /dev/rdsk/c0t3d0s5 | (cd
/backup;ufsrestore xf -)

in this way, the whole contents on c0t3d0s5 have been
dumped onto c0t1d0s5.

method 2, backup the whole filesystems structure from
an old disk into data file on the new disk:
 
example:
 # mount /dev/dsk/c0t1d0s7 /backup
 # cd /backup; 
 # mkdir ufsbackup
 # ufsdump 0f /backup/ufsbackup/root.ufs
/dev/dsk/c0t3d0s0

Thanks everyone again.

Chunhang


______________________________________________________________________ 
Find, Connect, Date! http://personals.yahoo.ca
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From galjanpa at cmr.gov  Thu May 30 09:50:07 2002
From: galjanpa at cmr.gov (Paul Galjan)
Date: Thu, 30 May 2002 09:50:07 -0400
Subject: SUMMARY: Firmware updates for Sun E4000
Message-ID: <20020530095007.M34984@cmr.gov>

Turns out Sun likes to refer to the E4000/4500 as the E4x00, so it wasn't 
turning up in my searches of sunsolve.  

Many thanks to: 
Rick McKinney
Tom Hetrick
Kevin Boykin
Charlotte Ratliff
Mike Marcell

--paul

> -----Original Message-----
> From: Paul Galjan [mailto:galjanpa at cmr.gov] 
> Sent: Wednesday, May 29, 2002 3:18 PM
> To: sunmanagers at sunmanagers.org
> Subject: Firmware updates for Sun E4000
> 
> Oh wise ones:
> 
> Does anyone know where I can find the firmware upgrades for the E4000?
> 
> Many TIA,
> --paul
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From mikelist at sky.net  Thu May 30 11:32:39 2002
From: mikelist at sky.net (Mike's List)
Date: Thu, 30 May 2002 10:32:39 -0500 (CDT)
Subject: SUMMARY: multi-threaded applications
In-Reply-To: <Pine.GSO.3.96.1020529142026.14432A-100000@apollo.sky.net>
Message-ID: <Pine.GSO.3.96.1020530102621.20821A-100000@apollo.sky.net>

Use top or prstat -a and look at the LWP (lightweight processes) or the
THR (threads) column, if it's more than 1 it's multi-threaded.  However,
someone did mentioned that an application that is multi-threaded does not
necessarily means it's taking advantage of multiple cpus.


- Mike


On Wed, 29 May 2002, Mike's List wrote:

> How do/can you determine an application is multi-threaded or not?
> Is there a command execute to find out? ie. ldd, crle, etc. on an
> application and what to look for?
> 
> The latest bind 9.x or apache 2.x supposed to take advantage of multi-cpus
> servers, but how do you go about testing and determines that these software
> is actually working? or other applications for that matter?
> 
> 
> - Mike
> _______________________________________________
> sunmanagers mailing list
> sunmanagers at sunmanagers.org
> http://www.sunmanagers.org/mailman/listinfo/sunmanagers
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From pat.novak at po.state.ct.us  Thu May 30 14:29:44 2002
From: pat.novak at po.state.ct.us (Patrick Novak)
Date: Thu, 30 May 2002 14:29:44 -0400
Subject: SUMMARY: dns failover problem
Message-ID: <001b01c20807$f87844e0$1e01a8c0@pjn.inetservices.doit.state.ct.us>

Many thanks for your prompt and useful replies

Dan Astoorian, John Riddoch and Brett Lymn all provided the correct
information, and that is:

nslookup is flawed. it is a troubleshooting tool and doesn't use
the same resolver mechanism that is used when a normal
dns lookup is being performed. It uses only one server at a time-
the first one listed in resolv.conf.

as I was testing a failover configuration for our dns, it became
apparent that I should use something that mimics the dns lookup
process more closely.

"getent hosts some.box.name" was the appropriate command
and did produce the desired results. when named was running
on box 1.2.3.4 and getent hosts www.some.name was entered,
the appropriate ip address was returned. I killed the named pid
on 1.2.3.4 and getent hosts www.some.other.name came back
with the proper ip address.

Pat Novak


Original email:

>I have 2 dns's, joined by a hub, set up as master and slave.  both have
each
>other
>entered as default routers, and both have each other listed in the
>resolv.conf file as
> second nameserver entry, and themselves as the first.  The slave finds and
>updates
>from the master as expected.

>the box at ip 1.2.3.4 has resolv.conf entries of:

>domain some.domain.com
>nameserver 1.2.3.4
>nameserver 1.2.3.5

>and  box at 1.2.3.5 has:
>domain some.domain.com
>nameserver 1.2.3.5
>nameserver 1.2.3.4

>if,  on dns A,  i say "nslookup   some.box.name"      it resolves.
>if,  on dns A,  i say "nslookup   some.box.name   dns.B.ip"  it resolves.
>and vice versa

>however, if i test a failover on box A by killing named-pid, then do:
>"nslookup some.box.name"  i get no response and truss shows the
>process sleeping. i get the same result going the other way.
>has anyone else seen this,  or have i missed something obvious.


>tia
>Pat Novak
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From lu at pgc.nrcan.gc.ca  Thu May 30 15:33:52 2002
From: lu at pgc.nrcan.gc.ca (Yuan Lu)
Date: Thu, 30 May 2002 12:33:52 -0700 (PDT)
Subject: SUMMARY: 'niscat auto_direct' shows nothing and automount
  doesn't work
Message-ID: <200205301933.MAA15984@taseko.pgc.nrcan.gc.ca>

Many thanks to Paul Hybert. His reply hits the jackpot!
Indeed for those problem users, they didn't get authenticated.

However, I found another strange thing: nisdefaults showed that
some users have been authenticated on one system, but not on
another system, although both systems are authenticated
and all user accounts are managed by NIS+. The user can choose
any system to login. Their home directory will be automounted to
whatever system they logged in.

Any suggestions, please.

Here is Paul's reply. My original message is at the bottom.

---------------------------------------------------------
The problem is with either NIS+ credentials for users, or
permissions on the NIS+ table.
niscat -o NIS+_table.name     shows the permissions on that table

nisdefaults                   shows information about the user 
running that command

Use those two commands to see details (especially on nisdefaults,
look for something about "user is not authenticated" or something
close to that.

Your message was not clear: you mention user-defined NIS+ table
named auto_direct, but the error message you showed was for 
auto_mailhost.

Maybe the error is not with THAT particular table itself, but
with the table that references it.

Paul Hybert
prh at ece.iit.edu

--------------------------------------------------------------
> 
> Hi there,
> 
> I think I might have problems of NIS+ permission problem, but
> I am not sure. Here is my story:
> 
> System: 5.8 Generic_108528-14 sun4u sparc SUNW,Ultra-2
> 
> This is a NIS+ client. The root user can use nisls to list
> all NIS+ tables. It can use niscat to display the contents of
> system tables , e.g. auto_master, auto_home, passwd, etc. 
However
> using niscat to display the contents of the user defined 
> auto_direct tables always give nothing. All regular users
> have the same problem as the root user, except one account.
> This luck user can display the contents of all the tables.
> 
> I have tried to setup the exact .profile  and .kshrc files
> as the 'luck user' for other users in hope that the others 
> will be able to display the user-defined auto_direct table. 
> I have no luck.
> 
> Because the root account can't display the contents of the
> auto_direct table, when I run automount as root, I got the 
> following error message:
> 
> May 29 10:05:36 myhost automount[15257]: [ID 527529 
> daemon.warning] nis_list: NIS+ error Not found encountered on 
> name auto_mailhost.org_dir.mydomain.name.ca. in table 
> auto_mailhost.org_dir.mydomain.name.ca.'s path.
> 
> I have another ultra10 system with the same OS and patch level
> having the similar problem. On this system the root and some
> users have no problems using niscat for all NIS+ tables. 
However,
> some other users can't display the user defined table
> auto_direct. Again I tried to setup all users in the same way: 
> the same ksh, same .profile, and using the same global
> /etc/profile. I don't have any luck at all.
> 
> Does any one know what's going on here? I will summarize
> the response.
> 
> Regards,
> 
> Yuan Lu
> _______________________________________________
> sunmanagers mailing list
> sunmanagers at sunmanagers.org
> http://www.sunmanagers.org/mailman/listinfo/sunmanagers
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From srinivas at satyam.com  Fri May 24 00:47:49 2002
From: srinivas at satyam.com (srinivas at satyam.com)
Date: Fri, 24 May 2002 10:17:49 +0530
Subject: Summary: E250 hanging at boot
Message-ID: <7FF62A49079FD511B14400065B19EF1205C0BD@cpr.satyam.com>

the problem turned out to be a failed HDD. Though there was nothing in the
vfstab and it is not a part of the md device, it is causing the solaris to
wait for a very long time before booting. After removing the faulty disk the
system boots like charm...I am still confused about how could a failed SCSI
cause the system to give a long pause. Anyone has clues?

Rene Occelli's mail helped alot. Here are others who gave helpful
hints..thanx to all of ya..:)

Rahul Parasnis [rahul.parasnis at creditlyonnais.fr]
Matt D. Harris [mdh at mdh.si.edu]
Kai Riasol Gonzalez [riasol at web.de]
Dennis Martens [Dennis_Martens at health.qld.gov.au]

-----Original Message-----
From: Srinivas_Arella at satyam.com [mailto:Srinivas_Arella at satyam.com]
Sent: Tuesday, May 21, 2002 1:14 PM
To: sunmanagers at sunmanagers.org
Subject: E250 hanging at boot


Hi All,

we have just received an E250 from another place(which obviously was
operational). The machine just refuses to boot even on cdrom. All
diagnostics come out clean. I noticed that one out of the six disk slots in
empty ( take out??!!! ). The missing disk apears to be on disk6 ( and
presumably not a boot disk).

The machine seems to boot fine till printing the Sun OS header and copyright
information ( on CD or from boot disk). I am unable to contact the admin who
handled the machine previously. Any help in tracing the problem would be
great.


TIA,
Srinivas
************************************************************************** 
This email (including any attachments) is intended for the sole use of the
intended recipient/s and may contain material that is CONFIDENTIAL AND
PRIVATE COMPANY INFORMATION. Any review or reliance by others or copying or
distribution or forwarding of any or all of the contents in this message is
STRICTLY PROHIBITED. If you are not the intended recipient, please contact
the sender by email and delete all copies; your cooperation in this regard
is appreciated.
**************************************************************************
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers
************************************************************************** 
This email (including any attachments) is intended for the sole use of the
intended recipient/s and may contain material that is CONFIDENTIAL AND
PRIVATE COMPANY INFORMATION. Any review or reliance by others or copying or
distribution or forwarding of any or all of the contents in this message is
STRICTLY PROHIBITED. If you are not the intended recipient, please contact
the sender by email and delete all copies; your cooperation in this regard
is appreciated.
**************************************************************************
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From ra7839 at sbc.com  Tue May 28 17:34:49 2002
From: ra7839 at sbc.com (ABENOJA, ROBERT (PBIS))
Date: Tue, 28 May 2002 14:34:49 -0700
Subject: Summary: Expect Script problems
Message-ID: <A425E1DB7EB4854B84BDF7E93B79876C1E6492@casnrm1msgusr01.ITServices.sbc.com>

I want to thank all those who responded.  I had an error in my script below
(expect -r "$prompt" it should be expect "machinename%" - sorry about that).
Josh Glover's hint's helped fix the problem.


Have you tried adding a short sleep before sending the password? This 
has helped me before.

e.g.:

> #!/usr/local/bin/expect
> log_file -noappend "telnetoutput.txt"
> spawn telnet 138.10.12.144
> expect "ogin: "
> sleep 1; send "dude\r"
> expect "assword: "
> sleep1; send "dude\r"
> expect -r "$prompt"
> send "cat /var/adm/messages | grep keyword"
> send "exit\r"

Worth a shot, maybe.


-- 
Josh Glover <jmglov at incogen.com>

Associate Systems Administrator
INCOGEN, Inc.




Thank you,
Bob Abenoja
PBIS Tech Supoort - San Ramon CA
925.806.4782


Original Message:
################

Dear Sun Gurus,

I have just installed expect 5.25 along with tcl-8.0 
and tk-8.0 on a solaris 2.6 sparc 5.

I modified a script I saw posted over a month ago.  
Here is a simple copy of my script:

#!/usr/local/bin/expect
log_file -noappend "telnetoutput.txt"
spawn telnet 138.10.12.144
expect "ogin: "
send "dude\r"
expect "assword: "
send "dude\r"
expect -r "$prompt"
send "cat /var/adm/messages | grep keyword"
send "exit\r"


The script gets to the password part and stops.  The 
prompt is then returned, almost looking like it tried 
to insert the prompt for the password.  I know the 
password is correct and I also tried another account.  
Can an Expect script expert help me figure this out.

Thank you,
Robert Abenoja
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From david at healthinsite.gov.au  Thu May 30 18:11:42 2002
From: david at healthinsite.gov.au (Dave Leach)
Date: Fri, 31 May 2002 08:11:42 +1000
Subject: SUMMARY:  BASEDIR change from request script goes unnoticed..
Message-ID: <07f101c20826$fab1a200$4d96a8c0@healthinsite.gov.au>

Looks like somone else had a similar problem at around the same time I did
with this very situation:

Solution:

Problem solved by including "cat > $1 <<EOM" within the 'checkinstall'
script.
This is what we were missing and once it was added, our problem was solved.

Thanks to
Deborah Santamauro
Todd Urie
Serge Borkov
Fabrice Guerini

A few people suggested using a different answer file, however this puts the
onus back onto the administrator installing the package, which is not what I
was after.  The basedir needs to be done calculated by the package install
scripts.
thanks again.

dave.


The original question:

> ----- Original Message -----
> From: Dave Leach
> To: sunmanagers at sunmanagers.org
> Sent: Friday, May 24, 2002 11:16 AM
> Subject: BASEDIR change from request script goes unnoticed..
>
>
> Hi there,
>
> I'm trying to modify the BASEDIR environment within a request script (it
> could equally go in the checkinstall script) under Solaris 8.  From what I
> understand BASEDIR can only be modified from within request or
checkinstall
> and only from versions 2.5 and greater.
>
> I have tried just setting it and also exporting it, and I still end up
with
> the BASEDIR as defined in the pkginfo file.  I've also tried using
> CLIENT_BASEDIR as well, and don't seem to be having much luck.  Do I need
to
> use a parametic base directory?  I didn't think I'd need to...
>
> The request/checkinstall script simply contains:
>
> #!/bin/sh
> BASEDIR=/opt/HIroot/opt
> echo "INFO: Using ${BASEDIR} as base directory for install..."
>
> Output from pkgadd...
>
> # pkgadd -d /tmp HIstuff
> Processing package instance <HIstuff> from </tmp>
>
> HI 1.4
> (sparc) 1.4
> HAC
> ## Executing checkinstall script.
> INFO: Using /opt/HIroot/opt/ as base directory for install...
> Using </opt> as the package base directory.
> ## Processing package information.
>
>
> any ideas?  I'm banging my head against a brick wall :-)
>
> dave
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From akrasanth at velankani.com  Fri May 31 00:47:06 2002
From: akrasanth at velankani.com (AK Rasanth)
Date: Fri, 31 May 2002 10:17:06 +0530
Subject: SUMMARY: Problem with cronjob
Message-ID: <02053110170602.01147@rasanth>

Hi All,
Thanks for all who had responded to my query. 
the answer for that is
When you are editing a cronfile without using crontab -e , then u need to 
explicitly signal to crond . the best way is to use crontab -e , this will 
signal the crond automatically after u save the file.

once again I thank all those who responded
Regards,
Rasanth

> Hi All,
> I have a backup script  which is to run as a cron job .  for this I made an 
> entry into /var/spool/cron/crontabs/root  
> 0 0 * * * /app/AdNet/Web/bin/backup.sh
> 
> the log file /var/cron/log says the following :
> CMD: /app/AdNet/Web/bin/backup.sh
> >  root 6478 c Tue May 28 16:30:00 2002
> <  root 6478 c Tue May 28 16:30:00 2002 rc=1
> 
> and nothing is being written on the tape. 
> can somebody help me out of this problem pls.....
-- 
A.K.Rasanth
ph: 91-80-8522507 Ext 1626

--------------------------------------------------------------
Velankani Information Systems Ltd, Bangalore, India
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From shaywood at hurricaneseye.com  Fri May 31 07:00:30 2002
From: shaywood at hurricaneseye.com (Haywood, Steven)
Date: Fri, 31 May 2002 12:00:30 +0100
Subject: SUMMARY -  L1000 Tape library barcodes
Message-ID: <D7028A02FC25D6119BC40008C71EFE050902E6@lonmsx02.rsl-europe.com>

Hiya

Thanks to all for offers of the truetype font ;) I had it already though.

The info I needed came from David Glass - the tape label needs "*"
characters around it (ie, *Label-01*) - thanks a bunch :)

Cheers
Steven

> -----Original Message-----
> From: Haywood, Steven [mailto:shaywood at hurricaneseye.com]
> Sent: Friday, May 31, 2002 10:58 AM
> To: 'sunmanagers at sunmanagers.org'
> Subject: L1000 Tape library barcodes
> 
> 
> Hiya
> 
> Sorry, I know this is OT slightly, but does anyone know what 
> data needs to
> be put on a barcode for a tape for the L1000 library? I know 
> it's in 3 of 9
> format, but I don't know what information needs to go into 
> the bar code
> 
> Thanks
> Steven
> 
> 
> **********************************************************************
> The information in this email is confidential and may be legally 
> privileged. It is intended solely for the addressee. Access to 
> this email by anyone else is unauthorised. If you are not the 
> intended recipient, any disclosure, copying, distribution or any
> action taken or omitted to be taken in reliance on it, is 
> prohibited and may be unlawful.
> **********************************************************************
> _______________________________________________
> sunmanagers mailing list
> sunmanagers at sunmanagers.org
> http://www.sunmanagers.org/mailman/listinfo/sunmanagers
> 


**********************************************************************
The information in this email is confidential and may be legally 
privileged. It is intended solely for the addressee. Access to 
this email by anyone else is unauthorised. If you are not the 
intended recipient, any disclosure, copying, distribution or any
action taken or omitted to be taken in reliance on it, is 
prohibited and may be unlawful.
**********************************************************************
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



From alan at esu.edu  Fri May 31 15:53:51 2002
From: alan at esu.edu (Alan Angulo)
Date: Fri, 31 May 2002 15:53:51 -0400 (EDT)
Subject: SUMMARY: Software Companion default installation path
In-Reply-To: <Pine.GSO.4.10.10205301208460.23859-100000@phys-esu0>
Message-ID: <Pine.GSO.4.10.10205311535550.7013-100000@phys-esu0>

The initial question was:
------------------------
On Thu, 30 May 2002, Alan Angulo wrote:
> Is it possible to change the default installation path (/opt/sfw) used by
> the installer in the Software Companion CD (Sparc Solaris 8) ?

Answer:
-------
Not possible

Alternative:
------------
1) Let it install in /opt/sfw
2) move /opt/sfw /to/wherever/you/need/it
3) Create a soft link /opt/sfw pointing /to/wherever/you/need/it

Experiences:
-----------
-If you create the /opt/sfw soft link first, the installer deletes it and 
 creates a directory.
-use prodreg to install and/or uninstall

Thanks to:
----------
Doug Otto who mentioned to choose custom install but didn't work
Jed Dobson who recommended to not touch it
Tim Evans who recommended to do a pkgadd -a none but will possibly break
David Foster who recommended to create a soft link before installing
Mike Penny  who recommended to move it after installing or creating a
            prior link


-- Alan Angulo
Systems Administrator
Academic Computing
East Stroudsburg University
alan at esu.edu
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers