From M.Russell at iaea.org  Wed Sep  1 02:08:56 2004
From: M.Russell at iaea.org (M.Russell at iaea.org)
Date: Wed, 01 Sep 2004 06:08:56 -0000
Subject: SUMMARY: possible to recover "bad" tapes?
Message-ID: <C4C76CB4FCDF5242B7519682A4E88C0233920A@sm1.sg.iaea.org>

YES !

Many thanks to all the people who responded:  Eugene  ;>D , Pablo Panchig,
Omyma Mahmoud Abelhafeez, Russell Page, Frank Smith, Ryan A. Krenzischek, Bob
Rahe, Joe Fletcher, Hendrik Visage.

About half of the respondents suggested retensioning the "bad" tapes, and the
other half thought the problem was probably with the drive and that I should
replace it.

Since I seem to have a good set of tapes at the moment (I haven't gotten any
write errors for more than a month), I decided yesterday to try the first
solution first.  Extrabyte doesn't allow the "retension" option in the mt
command, but since the erase command takes the tape to the end and back, I
used the command:

mt -f /dev/rmt/0 erase

The "bad" tape I prepared in this way yesterday was used for last night's
backup and no write errors occurred.

Thanks to everyone!

Marian Russell
Vienna, Austria

# # # # # # #

Original message:

I have been performing daily backups of SUNBlade 100 files to an Extrabye
8505
XL using 8mm Sony tape cartridges, 112m and 160m, for about 3 months.
According to one source, these tapes should be usable up to 50 times, and
yet,
in my experience so far, I am getting Write errors like:

Aug 11 03:24:27 sgcp1 scsi: [ID 107833 kern.warning]
WARNING: /pci at 1f,0/pci at 5/scsi at 0,1/st at 3,0 (st10):
Aug 11 03:24:27 sgcp1   Error for Command: write
Error Level:  Fatal
Aug 11 03:24:27 sgcp1 scsi: [ID 107833 kern.notice]
Requested Block: 497                       Error Block: 497
Aug 11 03:24:27 sgcp1 scsi: [ID 107833 kern.notice]
Vendor: EXABYTE                    Serial Number:
Aug 11 03:24:27 sgcp1 scsi: [ID 107833 kern.notice]
Sense Key: Media Error
Aug 11 03:24:27 sgcp1 scsi: [ID 107833 kern.notice]
ASC: 0x3 (<vendor unique code 0x3>), ASCQ: 0x2, FRU: 0x0

These errors occur after one or two uses, and sometimes even occur with new
tapes!  I now have about ten tapes that are supposedly bad.  Are these tapes
really bad?  Is there some way I can efficiently check them to see if they
are
really bad?  Can they be repaired somehow by reformatting?  It seems such a
waste to throw them away.

Is the problem possibly with the tape drive?  We have another Extrabyte 8505
XL tape drive that is not being used.  Should I try replacing the tape drive?

Thank you in advance for any advice.

Marian Russell
Vienna, Austria

This email message is intended only for the use of the named recipient.
Information contained in this email message and its attachments may be
privileged, confidential and protected from disclosure. If you are not the
intended recipient, please do not read, copy, use or disclose this
communication to others. Also please notify the sender by replying to this
message and then delete it from your system.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From Stefan.Varga at unit.sk  Wed Sep  1 10:09:52 2004
From: Stefan.Varga at unit.sk (Stefan Varga)
Date: Wed, 01 Sep 2004 14:09:52 -0000
Subject: Summary:Re: vx60 console
In-Reply-To: <41348B19.3050009@unit.sk>
References: <41348B19.3050009@unit.sk>
Message-ID: <4134A02B.9040905@unit.sk>

Hi again,

a. Open the bootenv.rc /boot/solaris

or jumpstart

/% /vi/ ///path_to_boot_server//boot/solaris/bootenv.rc


b. Add the following lines to the bootenv.rc file:

setprop input-device ttyb

setprop output-device ttyb

c. init 6
d. connect to serial port

thats it.
Stefan


Stefan Varga wrote:

> Hi all,
>
> is there a possibility to redirect the console (output-device & 
> input-device) on vx60 with Solaris 9 X86 to ttya.
> Because when i run the boot interpreter i dont see devices.
>
> Thank you.
> Stefan
> _______________________________________________
> sunmanagers mailing list
> sunmanagers at sunmanagers.org
> http://www.sunmanagers.org/mailman/listinfo/sunmanagers



-- 
+---------------------------------------------+
| Stefan Varga            UNIT spol. s r.o.   |
| System Engineer         Teslova 43          |
| mobil:+421908 7760617   82102 Bratislava    |
| mailto:varga at unit.sk    Slovakia            |
| Sun Microsystems Enterprise system provider | 
+---------------------------------------------+ 
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From bshah at citadon.com  Fri Sep  3 13:27:28 2004
From: bshah at citadon.com (Bhavesh Shah)
Date: Fri, 03 Sep 2004 17:27:28 -0000
Subject: SUMMARY: CPU idle time in Solaris 8
Message-ID: <0FEA43415756E44AB368AC58C1A4F4F301F9DF6A@a4mail2k01.citadoncorp.com>

Hi Gurus,
Thanks a lot for clarifying. I got lots of responses and very much
thankful to you all. The list is very big but to mention a few:
Darren Dunham
DarrenProg
Bossdad1959
Tristan Ball
Wade A. Lich
David P

Thanks Guys you rules.
Summary is:
Vmstat does not take into account iowait while calculating cpu idle time
where as top and mpstat does. So whenever there is a high iowait, vmstat
and top will show different results.

My Original Question was:

Hi Gurus,

I have a basic question regarding CPU Idle time in Solaris 8. I am
running Solaris 8 on Ultra 10 Box.

When I checked the CPU idle time from top utility it shows 12% idle
which is correct but at the same time if I check it from vmstat utility
it shows 90% idle. Is it normal behavior or some issue with the U10 box?

Any help will be greatly appreciated.

TIA

B





procs     memory            page            disk          faults
cpu

 r b w   swap  free  re  mf pi po fr de sr dd dd f0 m0   in   sy   cs us
sy id

 0 0 0 2212456 185376 65 85 30  0  0  0  0  0  0  0 31 1177  989  408  4
4 91

 0 0 0 2212456 184568 34  0  0  0  0  0  0  0  0  0 29 1199  509  389  2
4 94

 0 0 0 2212456 183848 45  0  0  0  0  0  0  0  0  0 29 1215  594  395  2
5 93

 0 1 0 2212456 182864 68  0  0  0  0  0  0  0  0  0 41 1574  766  489  4
4 91

 0 0 0 2212456 181488 63  0 49  0  0  0  0  0  0  0 46 1700  776  524 12
4 84

 0 1 0 2212456 180296 58 21 17  0  0  0  0  0  4  0 41 1577  553  478  2
6 92



load averages:  0.20,  0.29,  0.27
16:04:38

78 processes:  75 sleeping, 2 running, 1 on cpu

CPU states: 12.7% idle,  2.0% user,  2.0% kernel, 83.3% iowait,  0.0%
swap

Memory: 1024M real, 163M free, 2048M swap free



  PID USERNAME THR PRI NICE  SIZE   RES STATE   TIME    CPU COMMAND

10197 oracle    11  58    0    0K    0K sleep   2:45  2.22% oracle

10592 oracle     1  58    0 2168K 1416K cpu     0:01  0.94% top

27087 oracle     1  58    0    0K    0K sleep   1:02  0.54% oracle

 5751 oracle     1  58    0    0K    0K sleep   8:23  0.31% oracle

 5628 oracle    28  29   10  175M   80M run     9:56  0.16% java

 5647 oracle    11  58    0    0K    0K sleep   2:14  0.05% oracle

 5745 oracle     1  58    0    0K    0K sleep   3:55  0.04% oracle

 5743 oracle     1  58    0    0K    0K sleep   0:35  0.02% oracle

 5641 oracle     1  58    0    0K    0K sleep   0:05  0.02% oracle

12306 oracle    11  52    0    0K    0K sleep   7:00  0.01% oracle

 2704 oracle     6  58    0    0K    0K sleep   0:58  0.01% tnslsnr

    1 root       1  58    0  864K  344K sleep   0:14  0.01% init

 5649 oracle    11  59    0    0K    0K sleep   1:43  0.00% oracle

 5723 oracle     1  59    0    0K    0K sleep   0:35  0.00% oracle

 4641 oracle    13  58    0    0K    0K sleep  64:19  0.00% oracle
_______________________________________________
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 ranggon at dacafe.com  Sun Sep  5 19:57:50 2004
From: ranggon at dacafe.com (ranggon at dacafe.com)
Date: Sun, 05 Sep 2004 23:57:50 -0000
Subject: Summary: Monitor Disk Space
Message-ID: <34759.210.187.44.34.1094428635.squirrel@cafemail.edacafe.com>

Hi,

Many thanks to :

Surbeck Glen GS-09 868 CS/SCMC
Schaper, Soeren
Roetman, Paul
Rai, Brajshyam
Schaper, Soeren
Marcelino Mata
Joseph.Gaude at gd-ais.com
John Christian
Heiden, John
Dominique Potvin
D E
Bradley Alan
"Dvhr, Markus ICC-H"
Fishwick, Chris

Your willingness to share the scripts and the URLs is highly appreciated.

http://www.nagios.org/
bigadmin
http://www.bb4.org/

And to Michael Horton, your messages is noted.
your messages is noted.

Kind regards,
Ranggon


-----------------------------------------
Stay ahead of the information curve.
Receive EDA news and jobs on your desktop daily.
Subscribe today to the EDA CafeNews newsletter.
[ http://www10.edacafe.com/nl/newsletter_subscribe.php ]
It's informative and essential.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From Paul.Clayton at intec.co.za  Mon Sep  6 09:09:36 2004
From: Paul.Clayton at intec.co.za (Paul Clayton)
Date: Mon, 06 Sep 2004 13:09:36 -0000
Subject: Summary Sunfire F4800 problems
Message-ID: <9290BC9741D67F43BDAAC84144FF870D033C1227@intecserver007.intec.co.za>

Ok, so lots of egg on face with the vfstab file. We got lost on the device
names, and did not see that the "no' to mount at boot time should have been
"yes"

As for the /etc/system file. This one is odd. Although the system was
rebooted after the /etc/system was read, the changes did not show up in
"sysdef" until they were acccessed. Ie: a database created.

regards

-----Original Message-----
From: Nathan Dietsch [mailto:njd at ndietsch.com]
Sent: Monday, September 06, 2004 3:02 PM
To: Paul Clayton
Subject: Re: Sunfire F4800 problems


Hello Paul,


>In the /etc/vfstab file we have long device file name mapped to a
directory.
>These appear to not be read at boot time, but can be mounted bye command
>line as "mount -a".
>An example of a line is thus.
>/dev/dsk/c6t60003BA4E855400041386C190005D034d0s5
>/dev/rdsk/c6t60003BA4E855400041386C190005D034d0s5       /bmd    ufs     -
>no      rw,largefiles
>  
>
It seems that you have the mount at boot flag (number 6) set to no, set 
it to yes and it will mount at boot time.

>Next question is the /etc/system file.
>
>Any setting we put into the file get ignored. 
>
>The system is configured as a domain.
>
>  
>
Have you rebooted the machine since making the change?

I hope this helps.

Regards,

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


From andy.crichton at abibuildingdata.com  Mon Sep  6 09:38:03 2004
From: andy.crichton at abibuildingdata.com (andy.crichton)
Date: Mon, 06 Sep 2004 13:38:03 -0000
Subject: SUMMARY: Can not reset NVRAM on sunfire880
Message-ID: <13373747539192@dougal.abi-uk.co.uk>

I was having a problem whereby I could not reset the NVRAM (lost RSC
password) . The system was not flashing the maintenance LED during boot for
me to use the power button pressing method (i.e. press it twice during the
flashing). 

Sun have told me this is a system board failure which prevents normal OBP
operation. 
Strangely, this is the second board which has failed during power-off in
this host!

Thanks to John England and Joe Fletcher for taking the time to respond. 
There was mention of an OBP issue in a particular version which prevented
the double power button press from working. 

Andy. 


__________________________
This email and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed. The contents are confidential and may be privileged.
Any views or opinions expressed are those of the sender and may not reflect the views or opinions of the company and the company accepts no liability in respect thereof.
If you have received this email in error please notify postmaster at abibuildingdata.com immediately.

______________________________________________________________________
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email 
______________________________________________________________________
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From enordbye at broadpark.no  Wed Sep  8 16:58:47 2004
From: enordbye at broadpark.no (Eivind Nordbye)
Date: Wed, 08 Sep 2004 20:58:47 -0000
Subject: SUMMARY: scstat -i is hanging
In-Reply-To: <413AF75F.1030502@broadpark.no>
References: <413AF75F.1030502@broadpark.no>
Message-ID: <413F7261.8010002@broadpark.no>

Nobody replied to this message. I did not find a sollution, so I 
reinstalled the cluster with SC3.0 5/02, which worked all right.

Regards,
Eivind Nordbye

Eivind Nordbye wrote:

> Hi
>
> My configuration:
> Recently installed two node cluster (V210 and V240) running SC3.1 
> (from Java Enterprise System 2004Q2) and Solaris 8 2/04. bge 
> interfaces used as interconnects. D2 as shared storage.
>
> My problem:
> scstat -i is hanging for several minutes before it lists up the ipmp 
> group on the local node only. A message in the log says it cannot talk 
> to pnmd on the other node. Problem is the same on both nodes. It does 
> not hang when other node is down.  I had the problem before I 
> configured ipmp groups as well. The pnmd daemon is running on both 
> nodes and I can telnet to its' port number on the interconnet 
> interfaces from one node to another. Latest patches have been installed.
>
> Please help. Will summarize.
>
> Regards,
> Eivind Nordbye
> _______________________________________________
> 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 eric at astro.wesleyan.edu  Thu Sep  9 14:43:06 2004
From: eric at astro.wesleyan.edu (Eric Williams)
Date: Thu, 09 Sep 2004 18:43:06 -0000
Subject: SUMMARY: Automount won't automount. Troubleshooting
 recommendations requested.
In-Reply-To: <vqo1xhew6be.fsf@corbeau.imag.fr>
References: <24B71032-00D1-11D9-A04B-000393798960@astro.wesleyan.edu>
	<vqo1xhew6be.fsf@corbeau.imag.fr>
Message-ID: <03D70480-0290-11D9-9E49-000393798960@astro.wesleyan.edu>

Thanks to all who made suggestions. Here is what caused the issue:

> I think the MacOS X NFS server does not allow subdirectories to be
> mounted. You should check its configuration for this.
>
> Best regards.
>
> -- 
> Nicolas

This is kind of a pain and I am not sure of the workaround if I want to 
keep using NIS with auto_home maps. I'll be talking to my Apple 
engineer about why this is and if it can be fixed. Supposedly Solaris 
can be setup to be a client to a OS X server and use LDAP maps from the 
OS X machine to define mount points and such. I am working on that too, 
but now I am wondering how home areas from the Mac will be mounted if 
you can't mount subdirectories. Ugh! This was supposed to be easier...

Original question:

> Eric Williams <eric at astro.wesleyan.edu> writes:
>
>> Solaris 7 running NIS. I currently have home areas on several disks on
>> the network. The partitions are shared to all other machines so that I
>> can maintain a auto_home file distributed through NIS and have home
>> areas automount. I have recently added a Mac OS X server that shares a
>> large partition of home areas with NFS. When I add users to auto_home
>> with home areas on this share and try to login and connect  I get a
>> permission denied error. I can cd to this share through the /net mount
>> point as any user and see all the home areas so I know that NFS is
>> working and sharing is OK.
>>
>> Anyone have some troubleshooting tips as to why I can't automount home
>> areas from this share?
>


----
Eric Williams
Systems and Facilities Manager
Van Vleck Observatory - Wesleyan University Astronomy
Office: 860 685 3664
Cell: 860 670 3494
AIM: radvelman
http://www.astro.wesleyan.edu/~eric
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From hirdejit at gmail.com  Fri Sep 10 01:23:17 2004
From: hirdejit at gmail.com (Hirdejit Singh)
Date: Fri, 10 Sep 2004 05:23:17 -0000
Subject: SUMMARY: Problem in Recovery with Networker
In-Reply-To: <200409100027.RAA31191@redwood.taos.com>
References: <b50ff83404090904004481949@mail.gmail.com>
	<200409100027.RAA31191@redwood.taos.com>
Message-ID: <b50ff8340409092222373fdbc4@mail.gmail.com>

Hi,

Thanks for all those who replied.
When we save from the Cilent side , the Index are not saved in Server
but are written to tape. Since I was aware of teh Save Set ID, I
recovered teh Indexes with Scanner command.

However prior to running scanner, I moved the earlier client index to
other location and created a new index driectory in
/nsr.index/clientname/. And coomand nsrck -c clientname was used for
this.

Now on client while recovering I am able to see all files that I
backed up for mapped drives.

Regards


Original Post
> >
> > Hi,
> >
> > I am facing some problem in Networker while ecovering data. I have
> > mapped certain folders of other Windows machines on one of my Win2K
> > machine which is Networker Client.
> > Netwroker server is running on Solaris machine.
> >
> > Now to take backup of these mapped drives, I use following command
> > from Windows Client
> >
> > save  -s server -c client -I a.aa -b pool -L
> >
> > where a.aa is file containing the list of all mapped drives.
> >
> > Now  when I am trying to recover the files, it says the same is not in
> > index.I am not able to recover.
> > On server if I run mminfo command adn do a listing for the day I had
> > backed up this client I get following
> >
> >
> > gur/sun/1of5/sep0604/dc gursrv020 09/06/04 11:05:28 2000 MB 121881 cb
> > \\GURSRV025\WEBSPHERE\
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From Francois.Bousquet at cgi.ca  Fri Sep 10 13:09:10 2004
From: Francois.Bousquet at cgi.ca (Bousquet Francois)
Date: Fri, 10 Sep 2004 17:09:10 -0000
Subject: SUMMARY: syslog filter howto ?
Message-ID: <1F85129C88B3714185C912F3AF47605012A25316@nc1981exc03.banquelaurentienne.ca>

I found it before getting a answer :

slapd is logging on local4.debug and local4.debug by default so I had to
configured my syslogd that way :

# cat /etc/syslog.conf
*.debug;local4.none             /var/adm/messages
*.debug         @loghost
*.emerg         *

local4.info;local4.debug        /var/adm/slapd




-----Original Message-----
From: Bousquet Francois 
Sent: September 10, 2004 11:01 AM
To: Mailing List - SunManagers (E-mail)
Subject: syslog filter howto ?


Hi SunManagers,

syslogd is running on my Solaris 7 box.

I have an OpenLDAP slapd configured and running (starting line :
/soft/openldap/libexec/slapd -h "ldaps:/// ldap:///")

/etc/syslog.conf is set to this :

# cat /etc/syslog.conf
#
*.debug         /var/adm/messages
*.debug         @loghost
*.emerg         *
#


The problem is that slapd is logging a lot on the syslogs and I would like
to filter stuff with the syslod and to put it in another log ex :
/var/adm/slapd

But syslogd doesn't seems to have other filter types except the facility or
level.

How can I filter slapd logging ?
How can I know which facility.level a log to syslog is ?

Example of current syslog :
Sep 10 11:09:37 ub1981wfx01 slapd[27111]: conn=6 op=3 SRCH attr=uid
userPassword uidNumber gidNumber cn homeDirectory loginShell gecos
description objectClass
Sep 10 11:09:37 ub1981wfx01 slapd[27111]: conn=6 op=3 SEARCH RESULT tag=101
err=0 nentries=0 text=
Sep 10 11:09:51 ub1981wfx01 slapd[27111]: conn=9 fd=21 closed
Sep 10 11:09:51 ub1981wfx01 slapd[27111]: conn=10 fd=22 closed
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From kabinet at inf.u-szeged.hu  Fri Sep 10 13:10:58 2004
From: kabinet at inf.u-szeged.hu (kabinet at inf.u-szeged.hu)
Date: Fri, 10 Sep 2004 17:10:58 -0000
Subject: SUMMARY: bugid search
Message-ID: <4141E058.40706@inf.u-szeged.hu>

Hi Managers!

My original question was:
Where can I find description for bug 5077968? Are there any search pages 
for Solaris bugid's?

I had to register at SunSolve using my support contract number for 
full-search capabilities.

Thanks all,
Krisztian
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From peter at ashlyn.co.uk  Mon Sep 13 05:29:54 2004
From: peter at ashlyn.co.uk (Peter Stokes)
Date: Mon, 13 Sep 2004 09:29:54 -0000
Subject: SUMMARY - vold automatically starts AFTER system is running!!
Message-ID: <1095067658.1120.24.camel@sunsys.localdomain>

Hi All

Thanks to Dan O'Callaghan who hit it.

The ProDVD license had expired and needed a new one. Still do not know why it restarts vold though.....

Peter

Hi All

I have a curious problem with vold.

My U10 Sol 9 system (patched) has run for some time without a problem
and I have used cdrecord to write to my DVD-R's for about 1 year now. To
do this I simply take the vold down using the /etc/init.d/volmgt stop
commend and running cdrecord. I recently had to reconfig using devfsadm,
temporarily replace the CD with a replacement backup disk, copy data and
then replace the CD, reconfig's (devfsadm) and then all appeared fine.

Now however when I stop vold and issue the cdrecord commend to start the
write to the DVD, I get the cdrecord preamble, then

Alarm Clock
Volume management started

on the screen and the cdrecord can no longer get to the DVD.

Why is vol mgt starting of its own accord?

I can find no reference to thsi with Google etc. Any help would be
appreciated.

Thanks

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


From zzassgl at zoe.mcc.ac.uk  Mon Sep 13 07:08:47 2004
From: zzassgl at zoe.mcc.ac.uk (Geoff Lane)
Date: Mon, 13 Sep 2004 11:08:47 -0000
Subject: SUMMARY: Sol9 / LDAP / importing user info
In-Reply-To: <20040908094420.GA6362@zoe.mcc.ac.uk>
References: <20040908094420.GA6362@zoe.mcc.ac.uk>
Message-ID: <20040913110825.GA4376@zoe.mcc.ac.uk>

Everybody who wasn't away from their office pointed out that I should be
using ldapaddent.

I also found Sun Document ID 67966 which while not exactly addressing my
problem did provide some useful pointers.

Of course, the Solaris 9 System Administration Guide: Naming and Network
Directory Services (DNS, NIS and LDAP) is also useful.

My big mistake was starting with the LDAP documents in Sun's BluePrints
Collection.

Thanks,

On Wed, Sep 08, 2004 at 10:44:20AM +0100, Geoff Lane wrote:
> For a new Solaris 9 cluster project I want to use LDAP to manage the user
> information.  I've got the Sun 5.2 ldap server installed but now I'm stuck.
> 
> How do I import existing user information (in the form of /etc/passwd, group
> and so on) into the ldap database?
> 
> I've been trying to use dsimport but with zero success - possibly related to
> the fact that I can't find any current documentation for the utility (I've
> got Tom Bialaski's Sun BluePrint from 2001 but it seems that things have
> changed since then; no man page for dsimport is available in the 5.2 install.)
> 
> I'd really like an Idiots Guide to the process - understanding can come
> later when there is time - but pointers to any current documentation for
> dsimport would be welcome.
> 
> Thanks,
> Summary will follow.
> 
> 
> -- 
> | Geoff. Lane | Manchester Computing | Manchester | M13 9PL | England |
> 
> "Bother", said Pooh, as his LAN manager went crackers
> _______________________________________________
> sunmanagers mailing list
> sunmanagers at sunmanagers.org
> http://www.sunmanagers.org/mailman/listinfo/sunmanagers

-- 
| Geoff. Lane | Manchester Computing | Manchester | M13 9PL | England |

"Bother", said Pooh, as he fell into the nitric acid bath
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From Laurence.Moughan at aerlingus.com  Mon Sep 13 08:00:03 2004
From: Laurence.Moughan at aerlingus.com (Laurence Moughan)
Date: Mon, 13 Sep 2004 12:00:03 -0000
Subject: SUMMARY - issues with2  cross connected 3310 scsi array to 2 V240's
Message-ID: <s1459bdb.089@GWIADom>

My Chl3 on each 3310 was configured on a scsi id on controller 2 - I
only have one controller

access 3310 via console and reconfigured

DOH

.......................................................

Hi All,

I have connected two 3310 scsi arrays to 2 v240's.
I have build one raid 5 disks on the arrays constsing of the only 5
disks in each.

I have cabled up as single channel,

I have cabled up ch1 to node 1, ch3 to node 2  on each array.


from obp i can see

the disk of 3310 on ch1 - the array only of the 3310 on ch3  - on each
server.

when devfsadm;disks,

i can see then in format only the raid 5 disk on the connected 3310 on
chn1 - but not the "remote" disks on chn3


Anyone any ideas,  I only see mention also of the "local " chn 1 device
in messages as below,

pci at 3scsi@4 is the chn1 device - connected !
pci at 2scsi@4 is the ch3 device that i cannot see.

I have configured scsi inititator via an nvram file as per blueprintes
doc as below also ( on one machine only )


Sep 13 09:34:12 obeora1 scsi: [ID 365881 kern.info]
/pci at 1e,600000/pci at 2/scsi at 4 (qus0):
Sep 13 09:34:12 obeora1 pci_pci: [ID 370704 kern.info] PCI-device:
scsi at 5, qus1
Sep 13 09:34:12 obeora1 genunix: [ID 936769 kern.info] qus1 is
/pci at 1e,600000/pci at 2/scsi at 5
Sep 13 09:34:12 obeora1 scsi: [ID 365881 kern.info]
/pci at 1e,600000/pci at 2/scsi at 5 (qus1):
Sep 13 09:34:12 obeora1 pci_pci: [ID 370704 kern.info] PCI-device:
scsi at 4, qus2
Sep 13 09:34:12 obeora1 genunix: [ID 936769 kern.info] qus2 is
/pci at 1e,600000/pci at 3/scsi at 4
Sep 13 09:34:12 obeora1 scsi: [ID 365881 kern.info]
/pci at 1e,600000/pci at 3/scsi at 4 (qus2):
Sep 13 09:34:12 obeora1 pci_pci: [ID 370704 kern.info] PCI-device:
scsi at 5, qus3
Sep 13 09:34:12 obeora1 genunix: [ID 936769 kern.info] qus3 is
/pci at 1e,600000/pci at 3/scsi at 5
Sep 13 09:34:12 obeora1 scsi: [ID 365881 kern.info]
/pci at 1e,600000/pci at 3/scsi at 5 (qus3):
Sep 13 09:34:12 obeora1 scsi: [ID 193665 kern.info] sd61 at qus2:
target 0 lun 0
Sep 13 09:34:12 obeora1 genunix: [ID 936769 kern.info] sd61 is
/pci at 1e,600000/pci at 3/scsi at 4/sd at 0,0
Sep 13 09:34:12 obeora1 scsi: [ID 365881 kern.info]    
<SUN-StorEdge3310-0325 cyl 34746 alt 2 hd 127 sec 64>

/nvramsettings
probe-all
cd /pci at 1e,600000/pci at 3/scsi at 4
6 " scsi-initiator-id" integer-property
device-end
cd /pci at 1e,600000/pci at 2/scsi at 4
6 " scsi-initiator-id" integer-property
device-end
install-console
banner

#eeprom
use-nvramrc?=true
nvramrc=nvramrc=probe-all
cd /pci at 1e,600000/pci at 3/scsi at 4
6 encode-int " scsi-initiator-id" property
device-end
cd /pci at 1e,600000/pci at 3/scsi at 5
6 encode-int " scsi-initiator-id" property
device-end
install-console


Please advise if anyone can see the obvious !

Thanks

Laurence



For low fares and great deals on hotels, car hire and travel insurance visit http://www.aerlingus.com
of any action in reliance upon, this information by persons or entities
other than the intended recipient is prohibited.If you have received
this email in error please notify the sender immediately and delete
the material.
*******************************************************************************
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From benoit.audet at carra.gouv.qc.ca  Tue Sep 14 08:38:56 2004
From: benoit.audet at carra.gouv.qc.ca (=?ISO-8859-1?Q?Beno=EEt=20Audet?=)
Date: Tue, 14 Sep 2004 12:38:56 -0000
Subject: SUMMARY : xntpd exiting on startup with a strange log
  message...
Message-ID: <s146ade4.070@SG01.carra.gouv.qc.ca>

First of all, I would like to give a particular thanks to: John Hilger,
Dietmar Swoboda, Keith Jenkinson, Wallace MA Jeffrey Lai, Andy Scutt, and
Thorsten Clever...   ...thanks for having replied *automatically* to my
message that you were in vacation: that helped me a LOT! ;-)

Seriously, I would like to thank, in the first place, Neezam Haniff, who
pointed to me that I wasn't using "vmstat" correctly.  Mr. "darrenprog" did it
as well...  I had to let it go more longer to see the *real* stats.  In fact,
the first output, when usint "vmstat" without arguments, gives an average
since the system was rebooted...  ...not that usefull in my case!  So, I took
a bit of humility (ahem!) and did back my homeworks and found that I was short
in memory usage on the system...

Ok, I know, I know, it was pretty clearly mentionned in the log files, but I
wasn't suspecting it to be a *real* memory problem...   ...common, NTP
shouldn't need that much memory, isn't?  But I tried it back this morning,
with a little more memory left on the system, and guess what?  It passed
perfectly.  So, this was clearly a memory issue.

Once again, thanks!  Following is my original post...

Hi fellow managers,

I have a Solaris 8 running a Oracle 8i database, and that was running xntpd
without a problem.  I had to change the NTP servers today and everything went
fine on other servers, except this one.  As soon as I restart the xntpd
daemon, these entries are logged in "/var/adm/messages":

Sep 13 13:16:35 uprb01 ntpdate[24966]: [ID 558275 daemon.notice] adjust time
server 172.26.2.34 offset 0.125731 sec
Sep 13 13:16:35 uprb01 ntpdate[24966]: [ID 558275 daemon.notice] adjust time
server 172.26.2.34 offset 0.125731 sec
Sep 13 13:16:37 uprb01 xntpd[24968]: [ID 702911 daemon.notice] xntpd 3-5.93e
Mon Sep 20 15:47:11 PDT 1999 (1)
Sep 13 13:16:37 uprb01 xntpd[24968]: [ID 702911 daemon.notice] xntpd 3-5.93e
Mon Sep 20 15:47:11 PDT 1999 (1)
Sep 13 13:16:37 uprb01 xntpd[24968]: [ID 904279 daemon.error] mlockall():
Resource temporarily unavailable
Sep 13 13:16:37 uprb01 xntpd[24968]: [ID 904279 daemon.error] mlockall():
Resource temporarily unavailable
Sep 13 13:16:37 uprb01 xntpd[24968]: [ID 220665 daemon.error] No more memory!
Sep 13 13:16:37 uprb01 xntpd[24968]: [ID 220665 daemon.error] No more memory!

Of course, the daemon is not running after occurrence of these log entries...
I double checked the config files (/etc/inet/ntp.conf), I put back the old
one, checked on other servers and everything seems to be ok.

However, I looked at the memory usage on the server, with "vmstat", and I got
this:

server-name# vmstat
procs     memory            page            disk          faults      cpu
r b w   swap  free  re  mf pi po fr de sr m0 m1 m2 m3   in   sy   cs us sy
id
0 0 1 4722144 984656 0   0  0  0  0  0  0  5  4  4  0 4294967196 0 0 -4 -3
-141

Is this normal the negative values in the "cpu" columns?  I never saw that
before...

I browsed the web for something, by found nothing really relevant...
...only
one person who had this problem in july 2002, but no summary was posted back.

This obvious that I'll post a summary, as soon as I get it fixed!!  ;-)
Thanks in advance and have a nice day.

_______________________________________________________________
  Benoit Audet
  Conseiller - TELUS solutions d'affaires

  Commission administrative des rigimes de retraite et d'assurances (CARRA)
  Service de l'assistance et des technologies (SAT)
  Til : 418.528.1303  (CARRA)
  benoit.audet at carra.gouv.qc.ca
  benoit.audet at telussa.com
_______________________________________________________________
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From rangarita at telcel.net.ve  Tue Sep 14 09:30:53 2004
From: rangarita at telcel.net.ve (Rafael Angarita)
Date: Tue, 14 Sep 2004 13:30:53 -0000
Subject: SUMMARY: DNS Generic
Message-ID: <4146F263.80606@telcel.net.ve>

    Thanks A. Wooden,  C. Pinnock  and R. Kulawiec for your answers.

    The directive I was looking for is $GENERATE

    Original Question:

> I think there is a directive in bind to allow this and avoid to 
> declare each name explicitly, but I don't find it in the 
> documentation... it's something like "generic" I think... but don't know 
> the syntax...



    Below one of the answers:

    1. You're probably looking for $GENERATE, I think.

    2. Don't do it (the naming) this way. Do it this way:

    broadband-0001.dynamic.ourdomain.com
    broadband-0002.dynamic.ourdomain.com

    i.e. put all the end-user allocations (especially if they are
    cable, dialup or DSL using DHCP or similar) in a subdomain, and
    naem the subdomain something that clearly identifies as such.
    Other examples:

    broadband-0001.cablemodem.ourdomain.com

    broadband-0001.dhcp.ourdomain.com

    broadband-0001.dialups.ourdomain.com

    broadband-0001.dsl-users.ourdomain.com

    Why? Because that way you will make it much easier for people
    doing DNS and/or subdomain anti-spam blocking to list only those
    and not your entire domain.

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


From carlos_sevillano at ureach.com  Tue Sep 14 10:08:07 2004
From: carlos_sevillano at ureach.com (Carlos Sevillano)
Date: Tue, 14 Sep 2004 14:08:07 -0000
Subject: SUMMARY: void main(int argc, char *argv[]) question
Message-ID: <200409141407.KAA31903@www20.ureach.com>

Thanks to All who replied:

Casper Dik <casper at holland.sun.com>
"Thomas M. Payerle" <payerle at physics.umd.edu>
"Denes, Daniel \(BG-IT 51\)" <daniel.denes at bankgesellschaft.de> 
"Bill R. Williams" <brw at etsu.edu>
DAUBIGNE Sebastien - BOR <sebastien.daubigne at atosorigin.com>
"Meder Kydyraliev" <meder at areopag.net>
"Keith Smith" <keith at smith.net>
"Rafael Marcelino Koike" <r.koike at terra.com.br>
McEnroe <mcenroe at dli.ernet.in>
Rich Teer <rich.teer at rite-group.com>


The consensus is that the use of void on main is wrong (The book
is wrong), The use of "void main(int argc, char *argv[])"
is not part of the Standard and the fuction prototype should
return and int.  See answers below:
 


Casper Dik <casper at holland.sun.com>
>Solaris 2.8
>GCC 2.95.3  "SMCgcc"
>jedit, ddd, gdb
>
>In trying to compile some old test Unix C programs... I was not
>able to compile the programs.  After a while of tying to debug
>the programs that were of format:
>
>void main(int argc, char *argv[])
>or
>
>void
>main(int argc, char *argv[])

That's not legal according to the ANSI C standard; it also
resulted in bad exit code if main() falls of it end.

>failed to compile with errors on main:
>
># gcc -pg prog03.02.c -o prog032
>prog03.02.c: In function `main':
>prog03.02.c:15: warning: return type of `main' is not `int'
>
>The program compiled fine by removing "void" from the function
>main.  Why do those old UNIX programs used void ... it must
have
>been valid at some point ... there are quite a lot of such
>programs in Unix books ei Practical Unix programing, Rago Unix
>System Programming, Oreilly Unix System Programming etc.

You should replace "void" with "int" to be forward compatible
with C99.

The programs are wrong; there are only two acceptable
forms of main:

int main(void)
int main(int argc, char *argv[])

>What was the use of "void" as originally intended (granted it
>seems not work now days).  Did the language evolved (I tried
>-ansi on compiling trying with the same type of errors).  There
>is a lot of usage of void in the Unix programming literature.  

It was intended solely because the programmers were lazy and
wanted
to shut lint up about missing return statements in main.

>My question is in a nutshell can someone explain the usage
where
>it is ok to use void () and where it is not keeping in mind
>there a quite a lot of function prototypes that use void  ei:
>
>main(void)

This is ANSI C for "main doesn't expect arguments" and returns
(implictely)
an int.

>void handler(int)

This is a function which does not return a valuie.

>void main(int argc, char *argv[]) 

The function main does not return a value; however, that is not
allowed
by the standard.

BTW, I think you should use a newer form of gcc.



"Thomas M. Payerle" <payerle at physics.umd.edu>
> What was the use of "void" as originally intended (granted it
> seems not work now days).  Did the language evolved (I tried
> -ansi on compiling trying with the same type of errors). 
There
> is a lot of usage of void in the Unix programming literature.
Although not a professional c/c++ programmer, to my knowledge
void is
still very much a part of the c/c++ language.  When used for a
function
return type, it means the function does not return anything
(e.g. like
the old Pascal distinction between subroutines (not returning
anything)
and functions (a subroutine with a return value).

What has apparently changed is the strictness with which you can
mess
with the main subroutine.  The main subroutine on Unix like
systems was
_ALWAYS_ supposed to return something (an error code, or 0 is
successful).
I believe older C compilers allowed you to declare main as void
meaning that
you were not going to bother to set the return code.  This was
likely just
a special hack put in for main, and perhaps it allowed for some
consistency
across platforms; I do not claim to understand the historical
reasoning.
It might also be that older versions of c allowed the use of
void as a
return type to mean something along the lines of "I'm returning
something,
but I don't know what type yet".  With recent emphasis on
security, that is
probably a bad usage and was eliminated.

But on Unix, main() should always have an integer return type. 
I believe
you are still able to avoid specifying the parameters to main as
long as
you do not use them, but not positive about that.

void is still valid as a function return type, but only when the
function
actually does not return anything, and I believe will generate
an error if
the function actually gives an argument to the return command. 
Since main()
must rturn something to the OS, void is not appropriate for
main().

So most likely all you need to do is change void to int for the
main() subs,
and chances are if you used void elsewhere it should work.


"Bill R. Williams" <brw at etsu.edu>
The standard/legacy definition of a C program main() function is
that it returns an 'int'.  That's the way it WAS, and AFAIK
that's still
the way it IS.

Whoever wrote those programs you are working on has violated the
most basic level of writing a program in C.

I have seen compiler/library definitions vary on what the
argument list for main()  is -- always 'int argc' and 'char
*argv[]' or some equivalent variation -- and some allow more
arguments beyond for accessing the environmental stack, etc. 
BUT in all cases 'main()' should ALWAYS return 'int'; that's
where the operating system gets the "exit code" or "$?".  NOTE:
that some programs call 'exit()' rather than just (implicitly)
'return()', but you still should always declare
'int main(...)'.

-- 


DAUBIGNE Sebastien - BOR <sebastien.daubigne at atosorigin.com> 
ANSI C requires main to return an int (status code).
GCC defaults to Ansi C, so it warns you that the main prototype
is not ANSI
compliant.

Note that it is just a Warning, not an error. 
Nevertheless, the program will still be compiled, so you can
just ignore this message.

Another solution is to change the compilation flag to disable
Ansi and use another standard (see manpage).


"Keith Smith" <keith at smith.net>
"void handler(int)" or "void main(int argc, char *argv[])" is
used to mean that the function does not return a value. The
warning that gcc was giving u means that the function "main"
must return a int. This is then used as the return status of the
program.

"void handler(vold)" is the same as "void handler()"
"int main(void)" is the same as "int main()"

() was used in K&R C as a function prototype so (void) was need
to show a function that doesn't pass a parameter. Any new
compiler will treat () and (void) as the same. More people, so
I'm told, use () than (void).



"Rafael Marcelino Koike" <r.koike at terra.com.br>

All programs must return an error code.  This is why you need to
declare int main( int argc, char *argv[]) if you don't want to
return the error/sucess code, just return 0


McEnroe <mcenroe at dli.ernet.in>
as per the ANSI C standard default return type of a function is
'int'

it is better to use int,...if u have any value to return...or
if u dont have any value to return while a function fnish
executing ..then u can use void.


Rich Teer <rich.teer at rite-group.com>
> void main(int argc, char *argv[])
> or
>
> void
> main(int argc, char *argv[])
>
> failed to compile with errors on main:
>
> # gcc -pg prog03.02.c -o prog032
> prog03.02.c: In function `main':
> prog03.02.c:15: warning: return type of `main' is not `int'
>
> The program compiled fine by removing "void" from the function
> main.  Why do those old UNIX programs used void ... it must
have > been valid at some point ... there are quite a lot of
such

No.  void main () has NEVER been valid.

> programs in Unix books ei Practical Unix programing, Rago Unix
> System Programming, Oreilly Unix System Programming etc.

Bad habits picked up by programmers using compilers that didn't
flag that error.  :-(

> My question is in a nutshell can someone explain the usage
where
> it is ok to use void () and where it is not keeping in mind

In a C program (as opposed to C++), it is NEVER valid to have
void main().  Other functions that you (or others write) can,
however, be void.  For example, signal handlers don't return
a value, so they're void.



Original Posting:

>From   Carlos Sevillano <carlos_sevillano at ureach.com>  Add
contact  
To   sunmanagers at sunmanagers.org 
 Subject   void main(int argc, char *argv[]) question      
 
Solaris 2.8
GCC 2.95.3  "SMCgcc"
jedit, ddd, gdb

In trying to compile some old test Unix C programs... I was not
able to compile the programs.  After a while of tying to debug
the programs that were of format:

void main(int argc, char *argv[])
or

void
main(int argc, char *argv[])

failed to compile with errors on main:

# gcc -pg prog03.02.c -o prog032
prog03.02.c: In function `main':
prog03.02.c:15: warning: return type of `main' is not `int'

The program compiled fine by removing "void" from the function
main.  Why do those old UNIX programs used void ... it must have
been valid at some point ... there are quite a lot of such
programs in Unix books ei Practical Unix programing, Rago Unix
System Programming, Oreilly Unix System Programming etc.

What was the use of "void" as originally intended (granted it
seems not work now days).  Did the language evolved (I tried
-ansi on compiling trying with the same type of errors).  There
is a lot of usage of void in the Unix programming literature.  

My question is in a nutshell can someone explain the usage where
it is ok to use void () and where it is not keeping in mind
there a quite a lot of function prototypes that use void  ei:

main(void)
void handler(int)
void main(int argc, char *argv[]) 

void
main(int argc, char *argv[])

There is also:

main()
int main(int argc, char *argv[]);

Carlos

________________________________________________
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 psalazar at hotmail.com  Tue Sep 14 18:09:52 2004
From: psalazar at hotmail.com (Pablo Salazar)
Date: Tue, 14 Sep 2004 22:09:52 -0000
Subject: SUMMARY: disabling keyboard suspend/shutdown.
Message-ID: <BAY18-F34zOwPi0fLwh0004ef66@hotmail.com>

Thanks for all responses,

I do it this:

Edit the file /usr/openwin/lib/speckeysd.map

Just uncomment:

#SunPowerSwitch                 -       $OPENWINHOME/bin/sys-suspend
#SunPowerSwitchShift            -       $OPENWINHOME/bin/sys-suspend -n

kill and restart speckeysd.

# kill -HUP PID-of-speckeysd

Regards,

Pablo Salazar

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

The new MSN 8: smart spam protection and 2 months FREE*
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From zzassgl at zoe.mcc.ac.uk  Wed Sep 15 07:48:34 2004
From: zzassgl at zoe.mcc.ac.uk (Geoff Lane)
Date: Wed, 15 Sep 2004 11:48:34 -0000
Subject: SUMMARY: LDAP passwords not working / sol 9
Message-ID: <20040915114808.GA27962@zoe.mcc.ac.uk>

Thanks for the replies.  In the end I deleted the non-working configuration
and started again.

I installed patch 112960-18 which became available after I started working
on ldap authentication.

I then followed to the letter the example in the Solaris9 edition of System
Administration Guide: naming and Network Directory Services (DNS, NIS, and
LDAP)

http://docs.sun.com/db/doc/817-4843/6mkbebdbo?q=dns+nis+ldap&a=view

making only the necessary changes for a different server.

After that, ldap worked without any problems.

-- 
| Geoff. Lane | Manchester Computing | Manchester | M13 9PL | England |

McDonalds hamburgers are made from 100% genuine clown meat.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From daniel.denes at bankgesellschaft.de  Wed Sep 15 10:43:52 2004
From: daniel.denes at bankgesellschaft.de (Denes, Daniel (BG-IT 51))
Date: Wed, 15 Sep 2004 14:43:52 -0000
Subject: SUMMARY: Solstice admuseradd fails
Message-ID: <3641E2ABFD3EDD48B09F33361B0B8ADADFD686@TZEX020.bgb.de>

unfortunately received 0 helpful replies to this apart from the apparently
unavoidable "Out of office" spam from
Ballesteros, Dave D [ dave.ballesteros at eds.com ]
Stott, Trevor  xt55302 [ Trevor_Stott at adp.com ]
Dietmar.Swoboda at HVBInfo.com
Keith Jenkinson [ KJenkinson at ironmountain.co.uk ]
Lai, Jeffrey [ Jeffrey.Lai at gs.com ]
Sherman Butler [ sherman.butler at gettyimages.com ]
MA, Wallace, FM [ Wallace.MA at rbos.com ]
Bill Heese [ bill_heese at conair.com ]
Naveen Gajan [ ngajan at charlestoncounty.org ]
Thorsten Clever [ tclever at pironet-ndh.com ]
(are these becoming more lately?)

The problem was not that the user was not in the appropriate group
(admingroup) but that for some reason, his credential hasnt been updated in a
while; for some reason the groups are also stored within the "cred" table so
the entry for the admin group was missing there. nisclient -co <username>
solved this.

dd

--
> -----Urspr|ngliche Nachricht-----
> Von: sunmanagers-bounces at sunmanagers.org
> [mailto:sunmanagers-bounces at sunmanagers.org]Im Auftrag von
> Denes, Daniel
> (BG-IT 51)
> Gesendet: Donnerstag, 9. September 2004 18:08
> An: sunmanagers at sunmanagers.org
> Betreff: Solstice admuseradd fails
>
>
> Hello List,
>
> a new colleague tries to add a user to our NIS+ domain. The
> home directory of
> the new user should reside on a different server. My
> colleague is member of
> the admin group. Still when running
> admuseradd -u ... -c "..." -d /export/home/newUser -g ... -m -k
> /home/skeleton/skel -s ... -x ns=NIS+ -x serv=... -x pw=CLR
> -x autohome=Y
> -home=Y -x cred=N -x domain=... NewUser
> (this command is for ages being assembled by a wrapper script
> so typos are out
> of question), i get the error
> ERROR: add user method failed: Security exception on host ...
> . USER ACCESS
> DENIED.
>
> The user identity (1234).... was received, but that user
> is not authorized to execute the requested functionality on
> this system. Is
> this user a member of an appropriate security group on this system?
>
> Searching for this error, all i get is one Sun document that
> suggests the user
> is not member of the admin group.
>
> Any ideas?
>
> TIA
>
> dd
>
> --
> _______________________________________________
> 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 Andrew.Hardy at marks-and-spencer.com  Wed Sep 15 10:46:00 2004
From: Andrew.Hardy at marks-and-spencer.com (Hardy, Andrew)
Date: Wed, 15 Sep 2004 14:46:00 -0000
Subject: SUMMARY: Help required writing a case statement
Message-ID: <362014B7228974459E9E43F23989AB0036C44C@MSHVSMNSEXP0001.mnsexchange.adroot.marksandspencer.com>

All,

Thanks to the following people for their unbelievably quick responses !!

Jim Vandevegt
Kevin Smith
Gary Paveza
Dale Hichert
John Julian
Reggie Beavers
Lucien Hercaud
Daniel Denes
Mary Dell

The reason it would not run, was down to my own stupidity !!  When I
specified the "read ANSWER", I had prefixed it with a $  (see below)
Once I removed the dollar sign, it ran okay. Also, I had put esac on the
end, but I forgot to paste it into my original email, but thanks for
pointing it out anyway.



print  -n -- "PLEASE CHOOSE TO EITHER RELEASE ALL DRIVES OR BY AN
INDIVIDUAL FRAME. ( FR1 / FR2 / ALL ) : > "

        read $ANSWER

case $ANSWER in

        ALL|all) for drive in $ALL_DRIVE_LIST
              do
                 echo ""
                 echo "****** Now releasing drive $drive ******"
                 echo ""
                /usr/openv/volmgr/bin/vmoprcmd -crawlreleasebyname
$drive
            done
        ;;
esac


Many Thanks for all your help

Andy

-----Original Message-----
From: sunmanagers-bounces at sunmanagers.org
[mailto:sunmanagers-bounces at sunmanagers.org] On Behalf Of Hardy, Andrew
Sent: 15 September 2004 15:08
To: sunmanagers
Subject: Help required writing a case statement


All,

I am writing a script which requires me to embed a for loop within a
case statement. I have had a go at doing it (sample of code below), but
this does not run. However, the for loop on its own does run okay.

Does anyone have any idea if this can be done & if so what's the best
way to write it ?


case $ANSWER in

        ALL) for drive in $ALL_DRIVE_LIST

do
	echo ""
	echo "****** Now releasing drive $drive ******"
	echo ""
		/usr/openv/volmgr/bin/vmoprcmd -crawlreleasebyname
$drive
done ;;


Thanks in advance.

Andy Hardy



	Andy Hardy
	Technical Specialist, Data Storage Services,
	Marks & Spencer PLC, 3 Longwalk Road,Room 1SW,
	Stockley Park, Uxbridge,Middesex,UB11 1AW.
	* 020 8718 5975
	* 07733 333324
* mailto:andrew.hardy at marks-and-spencer.com





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


Registered Office:
Marks and Spencer plc
Waterside House
35 North Wharf Road
London
W2 1NW

Registered No. 214436 in England and Wales.

Telephone (020) 7935 4422
Facsimile (020) 7487 2670

www.marksandspencer.com

Please note that electronic mail may be monitored.

This e-mail is confidential. If you received it by mistake, please let
us know and then delete it from your system; you should not copy,
disclose, or distribute its contents to anyone nor act in reliance on
this e-mail, as this is prohibited and may be unlawful.

The registered office of Marks and Spencer Financial Services PLC, Marks
and Spencer Unit Trust Management Limited, Marks and Spencer Life
Assurance Limited and Marks and Spencer Savings and Investments Limited
is Kings Meadow, Chester, CH99 9FB. These firms are authorised and
regulated by the Financial Services Authority.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


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


Registered Office:
Marks and Spencer plc
Waterside House
35 North Wharf Road
London
W2 1NW

Registered No. 214436 in England and Wales.

Telephone (020) 7935 4422
Facsimile (020) 7487 2670

www.marksandspencer.com

Please note that electronic mail may be monitored.

This e-mail is confidential. If you received it by mistake, please let us know
and then delete it from your system; you should not copy, disclose, or
distribute its contents to anyone nor act in reliance on this e-mail, as this
is prohibited and may be unlawful.

The registered office of Marks and Spencer Financial Services PLC, Marks and
Spencer Unit Trust Management Limited, Marks and Spencer Life Assurance
Limited and Marks and Spencer Savings and Investments Limited is Kings Meadow,
Chester, CH99 9FB. These firms are authorised and regulated by the Financial
Services Authority.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From Alsandy6746 at cs.com  Thu Sep 16 05:31:53 2004
From: Alsandy6746 at cs.com (Alsandy6746 at cs.com)
Date: Thu, 16 Sep 2004 09:31:53 -0000
Subject: Summary: Stupid Hardware Question-Ultra10
Message-ID: <15d.3f807754.2e7ab784@cs.com>

we have a calligraphix calligraphy system, almost like Inscribe. we are 
looking for a used bjc 800 printer or a used Inscribe system.

can you help?

al kadan


From SunManagers at lycos.co.uk  Thu Sep 16 07:20:51 2004
From: SunManagers at lycos.co.uk (SunManagers at lycos.co.uk)
Date: Thu, 16 Sep 2004 11:20:51 -0000
Subject: SUMMARY: log NFS requests
Message-ID: <1095333606002661@lycos-europe.com>

 Hi Managers,

Late Summary due to childbirth   :-)

Problem:-
> I have an application that is making an extremely large number of file
> reads to an NFS server. I would like to log (probably only for a brief
> period of time) which files/directories it is accessing. OS is Solaris 8.

Thanks to:

the man pages for truss and System Calls; Casper Dik; Rich Teer; Chris Pinnock.

Solution:
man nfs_share (suggested by many; I couldn't get this to work)

truss -t open,read,lseek -p $PID_OF_RUNNING_PROCESS (suggested by Chris). I found that 
truss -t open -o /tmp/trussoutput -p $PID_OF_RUNNING_PROCESS was sufficient to get the information required and keep it in a useable form.

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


From blymn at baesystems.com.au  Thu Sep 16 22:25:31 2004
From: blymn at baesystems.com.au (Brett Lymn)
Date: Fri, 17 Sep 2004 02:25:31 -0000
Subject: [SUMMARY] rotating squid logs using logadm
In-Reply-To: <20040916123419.GA56@baea.com.au>
References: <20040916123419.GA56@baea.com.au>
Message-ID: <20040917022447.GC194@baea.com.au>

Last night I posted the message quoted below.  I received the flood of
out of office messages along with a very helpful response from Bill R.
Williams.  Though Bill did not directly give me the answer he pointed out
how the linux logrotate worked with squid, from that I managed to brew
up the right logadm syntax:

squid -C 60 -a '/opt/local/squid/sbin/squid -k rotate; sleep 2; rm -f /logs/squi
d/*.log.0' -c -p 1d -t '$file.%Y%m%d' -z 0 '/logs/squid/{access,cache,store}.log

The logs get copied and truncated.  After that I tell squid to rotate the logs
and remove the redundant copy of the logs.  There is a small window of
log info we don't capture but I am not too concerned about that.

Thanks be to Bill.

Liberal application of the clue-stick goes to:

tclever at pironet-ndh.com
Wallace.MA at rbos.com   
sherman.butler at gettyimages.com
Andy.Scutt at jobpartners.com
John.Hilger at veritas.com
KJenkinson at ironmountain.co.uk
Dietmar.Swoboda at HVBInfo.com

Come on people, even outlook's lame mail filtering capabilities could have
saved this useless notification.

On Thu, Sep 16, 2004 at 10:04:19PM +0930, Brett Lymn wrote:
> Folks,
> 
> I brave the veritable deluge of out of office notifications (waves 
> cluestick on high at said persons) to seek help with logadm.
> 
> I have a squid proxy server and in the past I have run a script to rotate
> my squid logs around.  I wanted to use logadm in solaris 9 to do this since
> it looks to be a neater solution than having lots of little log rotate
> scripts.  The problem is that I cannot get the syntax right, when you
> ask squid to rotate it's logs it produces a file of the form access.log.0
> which is the file I want to rename, and compress.  The problem is that
> if I tell logadm to look for access.log.0 it says the file does not exist
> and bails out, the -N option suppresses the error message but no logs are
> rotated.  I am half tempted to a) log a bug report with Sun saying the
> presence of the log file should be checked _after_ the pre-rotate script
> is run or b) resort to scripting my own log rotation again.  Before I do
> either of these I thought I would throw myself upon the collective wisdom
> (and auto-responders...grrr) of this list.  This is the logadm entry
> for my squid rotation:
> 
> squid -C 60 -b '/opt/local/squid/sbin/squid -k rotate; sleep 240' -p 1d -t $file.%Y%m%d -z 0 '/logs/squid/{access,cache,store}.log.0'
> 
> I don't really want to shutdown squid every night just to rotate the logs.
> 
> Suggestions most welcome.
> 

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


From neeraj_rohliyan at rediffmail.com  Fri Sep 17 03:27:15 2004
From: neeraj_rohliyan at rediffmail.com (Neeraj Kumar Rohliyan)
Date: Fri, 17 Sep 2004 07:27:15 -0000
Subject: Summary: Time/Date problem in Netscape 4.15 Messaging server
Message-ID: <20040917072628.17450.qmail@webmail8.rediffmail.com>

  hi 
I got no response at all. I have solve the problem by setting The System
TIMEZONE Asia/Calcutta.


On Wed, 15 Sep 2004 Neeraj Kumar Rohliyan wrote :
>
Hi Gurus
>
>
>I have Netscape Messaging Server 4.15 installed on Solaris 2.6
.The problem i m facing is that my system TIME ZONE is GMT+5.When i send a
message it shows the 10 hours ahead on recieving end. when i read the header
of message it show the following time stamp
>
>ESMTP 9:00 -05:00
>
>HTTP
19:00 +5:00
>
>please suggest me what is the problem.
>
>
>thankx
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From sebastian.boeker at hsh-nordbank.com  Fri Sep 17 07:51:53 2004
From: sebastian.boeker at hsh-nordbank.com (sebastian.boeker at hsh-nordbank.com)
Date: Fri, 17 Sep 2004 11:51:53 -0000
Subject: SUMMARY: NFS Question: What server has mounted a NFS share? 
 [Virus checked HSH Nordbank]
Message-ID: <20040917115115.352DD1E79B@sunportal.sunmanagers.org>

Hi,
thank you very much for the immediately answers.

Simon Burra and  Daniel Denes suggests:
   /usr/sbin/showmount -a <HOST>

Angelo,  Andrew Hardy and Alan Bradley suggests
   /usr/sbin/dfmounts <HOST>


Both works fine and is exactly what I'm looking for.

Best regards
Sebastian Boeker

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

Sebastian Boeker
HSH Nordbank AG, OE 1375,
Decentralize Systems
Schlo_garten 14, D- 24103 Kiel
Email: sebastian.boeker at hsh-nordbank.com
Tel: 0431-900-12285
Fax: 0431-900-6-12285

----- Weitergeleitet von Sebastian Bvker/13/HSH Nordbank/DE am 17.09.2004
13:43 -----
|---------+----------------------------------->
|         |           sebastian.boeker at hsh-nor|
|         |           dbank.com               |
|         |           Gesendet von:           |
|         |           sunmanagers-bounces at sunm|
|         |           anagers.org             |
|         |                                   |
|         |                                   |
|         |           17.09.2004 12:57        |
|         |                                   |
|---------+----------------------------------->
  >--------------------------------------------------------------------------
----------------------------------------------------|
  |
|
  |       An:       <sunmanagers at sunmanagers.org>
|
  |       Kopie:
|
  |       Thema:    NFS Question: What server has mounted a NFS share?  [Virus
checked HSH Nordbank]                             |
  >--------------------------------------------------------------------------
----------------------------------------------------|




Hi,
I am looking for a way to display all servers who has mounted
a NFS share.

Server POLLUX shares a directory.
Entry in /etc/dfs/dfstab:
share -F nfs -o rw -d "ASCII-Expot" /export/ascii_export

And it works well.
Is there a possibility to find out on server POLLUX  on which servers this
NFS share is mounted?

Thanx for your help.

Best regards
Sebastian Bvker

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

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

Sebastian Bvker
HSH Nordbank AG, OE 1375,
Dezentrale Systeme
Schlo_garten 14, D- 24103 Kiel
Email: sebastian.boeker at hsh-nordbank.com
Tel: 0431-900-12285
Fax: 0431-900-6-12285

_____________________________________________________________________________

___________

Diese Nachricht kann vertrauliche Informationen enthalten. Sollten Sie
nicht der vorgesehene Empfdnger sein, so bitten wir Sie, den Absender
unverz|glich zu informieren und die E-Mail zu lvschen. Jeder unbefugte
Zugriff oder unbefugte Weiterleitung, die Fertigung einer Kopie, die
Vervffentlichung oder sonstige in diesem Zusammenhang stehende Handlung ist
untersagt. Da wir nicht die Echtheit oder Vollstdndigkeit der in dieser
Nachricht enthaltenen Informationen garantieren kvnnen, schlie_en wir die
rechtliche Verbindlichkeit der vorstehenden Erkldrungen und Du_erungen aus.

This message may contain confidential information. If you are not the
intended recipient please promptly inform the sender and delete this email.
Any other unauthorized access or unauthorized forwarding, copy creation,
publication or any other action in this connection is prohibited. As we
cannot guarantee the genuineness or completeness of the information
contained in this message, the statements set forth above are not legally
binding.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



_____________________________________________________________________________
___________

Diese Nachricht kann vertrauliche Informationen enthalten. Sollten Sie
nicht der vorgesehene Empfdnger sein, so bitten wir Sie, den Absender
unverz|glich zu informieren und die E-Mail zu lvschen. Jeder unbefugte
Zugriff oder unbefugte Weiterleitung, die Fertigung einer Kopie, die
Vervffentlichung oder sonstige in diesem Zusammenhang stehende Handlung ist
untersagt. Da wir nicht die Echtheit oder Vollstdndigkeit der in dieser
Nachricht enthaltenen Informationen garantieren kvnnen, schlie_en wir die
rechtliche Verbindlichkeit der vorstehenden Erkldrungen und Du_erungen aus.

This message may contain confidential information. If you are not the
intended recipient please promptly inform the sender and delete this email.
Any other unauthorized access or unauthorized forwarding, copy creation,
publication or any other action in this connection is prohibited. As we
cannot guarantee the genuineness or completeness of the information
contained in this message, the statements set forth above are not legally
binding.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From d.matinyarare at piorin.gov.pl  Fri Sep 17 09:51:19 2004
From: d.matinyarare at piorin.gov.pl (David Matinyarare)
Date: Fri, 17 Sep 2004 13:51:19 -0000
Subject: SUMMARY: root file system
Message-ID: <6682A0A20A720942AC26337E5E5C03EA1435FA@www.iorin.gov.pl>

Thanks a lot to all those who spared their time to help me. Most of you seemed
to suggest file growth in the root file system and you where correct. A file,
0n (which I thought and beleave could have been linked the /devices) was
keeping growing in the /dev/rmt directory as 0n. This file was growing after
each backup taken and I am still wondering why and how could it happen. As it
is a different issue altogether, if there is a quick respond from anyone of
you, it is wellcome. At the moment I will cool down and look at the new mistry
carefully and as time permits.
Thanks again for all quick responds which I received.

My orginal request was as:

My root file system is full and I was trying to find out whether I could
extend it somehow, but I could not find a way to do so. Does anyone out
there
know how I can go about it without, backing the system to tape, reorganise
the
file system and restore the system. For a much clear understanding of where
I
have bungled I have included the df output and a cut off from the format
output. Links to appropriate documentation is well come.

Please note that my disk was fully partioned.



Filesystem            kbytes    used   avail capacity  Mounted on
/dev/dsk/c1t0d0s0    1988887 1968990       0   100%    /
/dev/dsk/c1t0d0s3    5046030 1644689 3350881    33%    /usr
/proc                      0       0       0     0%    /proc
mnttab                     0       0       0     0%    /etc/mnttab
fd                         0       0       0     0%    /dev/fd
/dev/dsk/c1t0d0s4    5046030  516898 4478672    11%    /var
swap                 15647968      40 15647928     1%    /var/run
swap                 15648272     344 15647928     1%    /tmp
/dev/dsk/c1t1d0s7    70592505 5230381 64656199     8%    /oracle
/dev/dsk/c1t0d0s5    5046030 2948396 2047174    60%    /opt
/dev/dsk/c1t0d0s7    41332195 4824315 36094559    12%    /export/home




Part      Tag    Flag     Cylinders         Size            Blocks
  0       root    wm    2415 -  2817        1.96GB    (403/0/0)     4100928
  1       swap    wu       0 -  2414       11.72GB    (2415/0/0)   24575040
  2     backup    wm       0 - 14086       68.35GB    (14087/0/0) 143349312
  3        usr    wm    4832 -  5838        4.89GB    (1007/0/0)   10247232
  4        var    wm    3825 -  4831        4.89GB    (1007/0/0)   10247232
  5 unassigned    wm    2818 -  3824        4.89GB    (1007/0/0)   10247232
  6 unassigned    wm       0                0         (0/0/0)             0
  7       home    wm    5839 - 14086       40.02GB    (8248/0/0)   83931648
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From ml at kcore.org  Mon Sep 20 01:50:25 2004
From: ml at kcore.org (Jan De Luyck)
Date: Mon, 20 Sep 2004 05:50:25 -0000
Subject: SUMMARY: Disable onboard hme0 on ultra5
In-Reply-To: <200409171829.36682.ml@kcore.org>
References: <200409171829.36682.ml@kcore.org>
Message-ID: <200409200749.04836.ml@kcore.org>

Thanks everyone who answered. There were two main suggestions:

1. unplumbing the interface using ifconfig. This way is unfortunately unusable 
for me, since there are too many configuration files that have the interface 
in question in them, and the machines should be identical if possible.

2. removing the entry from /etc/path_to_inst: this is wat we ended up doing, 
and which worked great: Solaris doesn't see the interface anymore, the 
'older' hme1 is now hme0 and everything works peachy. Just no reconfiguration 
boots ;p

Jan

On Friday 17 September 2004 18:29, Jan De Luyck wrote:
> Hello List,
>
> I'm currently having a bit of trouble with one of our ultra5 boxes - dead
> hme0. It's not possible to get it shipped for replacement, so I'm making
> due with two PCI NIC's.
>
> The onboard transceiver is fubar, doesn't work, receives nothing. Is it
> possible to deactivate this in some way so that Solaris doesn't see it
> either? This because we have a standard jumpstart server that I would like
> to be able to use also on this modified box....
>
> Thanks.
>
> Will summarize.
>
> Jan

-- 
You will be successful in love.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From russellpage at hotmail.com  Mon Sep 20 03:34:24 2004
From: russellpage at hotmail.com (Russell Page)
Date: Mon, 20 Sep 2004 07:34:24 -0000
Subject: SUMMARY: Notebook for support.
Message-ID: <BAY19-F27f3sllNY30k00010a56@hotmail.com>

Thanks to all who replied. The general consensus is a refurbished Thinkpad. 
No one mentioned any other brand. Do we know something other notebook users 
don't?

Chris (kingsqueak at kingsqueak.org) has this to say about USB-> serial 
adaptors: " ... the USB serial dongles are HORRIBLE for freezing up and also 
for sending a break at random points while in use."

I shudder to think what one of those would do to one of our production 
firewalls ($ 10^6 / hour financial traffic! ).

ttg (ttg at charter.net) reminded me to look at Ebay for a 240. A sudden flash 
of the blindingly obvious! I hadn't thought of that. Maybe I'll be able to 
buy my own system back!

Thanks again everyone!

-- Russell Page.

Veni, vidi, modus creavi.

_________________________________________________________________
Click here for the latest chart ringtones:  
http://ringtones.com.au/ninemsn/control?page=/ninemsn/main.jsp
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From jdunn at sefas.com  Mon Sep 20 03:50:53 2004
From: jdunn at sefas.com (John Dunn)
Date: Mon, 20 Sep 2004 07:50:53 -0000
Subject: SUMMARY: rdate sets wrong time
Message-ID: <000001c49ee6$7c561720$3e04050a@johnman>

Thanks for the many responses. The problem was, as many suggested, different
timezones on AIX and Solaris.
In the end setting TZ=GMT-1 on Solaris seemed to do the trick.



-----Original Message-----
From: Sandwich Maker [mailto:adh at an.bradford.ma.us]
Sent: Friday, September 17, 2004 4:24 PM
To: jdunn at sefas.com
Subject: Re: rdate sets wrong time


" From: "John Dunn" <jdunn at sefas.com>
"
" I am attempting to synhronise the time on my Solaris 8 server with a AIX
" server using xntpd. (I already synchronise a number of AIX boxes OK)
" However when I run
"
" rdate aixserver
"  it returns time that is 2 hours behind the time on the AIX server!.
"
" Any explaination?

different timezones?

if you're gonna use xntp, why rdate and not ntpdate?
________________________________________________________________________
Andrew Hay                                  the genius nature
internet rambler                            is to see what all have seen
adh at an.bradford.ma.us                       and think what none thought
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From martini at mrpeabody.llnl.gov  Mon Sep 20 12:29:01 2004
From: martini at mrpeabody.llnl.gov (Dave Martini 1)
Date: Mon, 20 Sep 2004 16:29:01 -0000
Subject: SUMMARY: Booting from the root mirror using DiskSuite PLUS
 FOLLOW UP QUESTION
Message-ID: <200409201625.i8KGPAv08103@raider.llnl.gov>

Thanks to the many replies I got on this!

First off, Solaris 9 comes with Disksuite in the OS so no need to install any additional
packages for it.

With Disksuite, You need n+1 state database replicas where n is half the total number of replicas
for the system to be able to boot up in multiuser mode.

With only 2 disks in the system, the original boot disk and the mirror, this can't
be achieved. 

So, there is a flag you can put in your /etc/system file which will tell the machine
to boot up even though there are only 50% state database replicas.

Here is the line to add to /etc/system

set md:mirrored_root_flag=1

After putting this line in my /etc/system I shut the system down and pulled the original boot
disk out and booted from the mirror and it came up in multi user mode just like it was 
supposed to. 

After putting the original boot disk back in and booting from it, the system came up ok
once again. I do notice now however from the metastat command that it's wanting me to 
issue a metareplace of the state database replicas from the original boot disk that I pulled
for my test even though the original boot disk is back in the system.

It would be nice if Disksuite would have recognized that the original boot disk is there
and that there are two copies of the state database replica there and fix/resync it on its own.

Since this is the original boot disk can I just execute the commands

# metareplace -e d8 c0t0d0s3   (c0t0d0 is the boot disk)
# metareplace -e d5 c0t0d0s1  
# metareplace -e d0 c0t0d0s0

According to the metareplace man page the -e does the resync automatically??????

Here's the disk assignments from the format command

AVAILABLE DISK SELECTIONS:
       0. c0t0d0 <SUN72G cyl 14087 alt 2 hd 24 sec 424>  <<<<<<<<<<<<<<<<< Original Boot disk
          /pci at 1d,700000/scsi at 4/sd at 0,0
       1. c0t8d0 <SUN72G cyl 14087 alt 2 hd 24 sec 424>  <<<<<<<<<<<<<<<<<<< Boot Mirror
          /pci at 1d,700000/scsi at 4/sd at 8,0

Here is how I have the 2 disks sliced up

Part      Tag    Flag     Cylinders         Size            Blocks
  0       root    wm       0 -  3018       14.65GB    (3019/0/0)   30721344
  1       swap    wu    3019 -  4629        7.82GB    (1611/0/0)   16393536
  2     backup    wm       0 - 14086       68.35GB    (14087/0/0) 143349312
  3        var    wm    4630 -  5636        4.89GB    (1007/0/0)   10247232
  4 unassigned    wm    5637 -  5638        9.94MB    (2/0/0)         20352
  5 unassigned    wm       0                0         (0/0/0)             0
  6 unassigned    wm       0                0         (0/0/0)             0
  7 unassigned    wm       0                0         (0/0/0)             0

partition> 

Here is the output from my metastat command and you will see some submirrors have a
"needs maintenantce" flag.

And my metastat output

# metastat
d8: Mirror
    Submirror 0: d6
      State: Needs maintenance 
    Submirror 1: d7
      State: Okay         
    Pass: 1
    Read option: roundrobin (default)
    Write option: parallel (default)
    Size: 10247232 blocks (4.9 GB)

d6: Submirror of d8
    State: Needs maintenance 
    Invoke: metareplace d8 c0t0d0s3 <new device>
    Size: 10247232 blocks (4.9 GB)
    Stripe 0:
        Device     Start Block  Dbase        State Reloc Hot Spare
        c0t0d0s3          0     No     Maintenance   Yes 


d7: Submirror of d8
    State: Okay         
    Size: 10247232 blocks (4.9 GB)
    Stripe 0:
        Device     Start Block  Dbase        State Reloc Hot Spare
        c0t8d0s3          0     No            Okay   Yes 


d5: Mirror
    Submirror 0: d3
      State: Needs maintenance 
    Submirror 1: d4
      State: Okay         
    Pass: 1
    Read option: roundrobin (default)
    Write option: parallel (default)
    Size: 16393536 blocks (7.8 GB)

d3: Submirror of d5
    State: Needs maintenance 
    Invoke: metareplace d5 c0t0d0s1 <new device>
    Size: 16393536 blocks (7.8 GB)
    Stripe 0:
        Device     Start Block  Dbase        State Reloc Hot Spare
        c0t0d0s1          0     No     Maintenance   Yes 


d4: Submirror of d5
    State: Okay         
    Size: 16393536 blocks (7.8 GB)
    Stripe 0:
        Device     Start Block  Dbase        State Reloc Hot Spare
        c0t8d0s1          0     No            Okay   Yes 


d0: Mirror
    Submirror 0: d1
      State: Needs maintenance 
    Submirror 1: d2
      State: Okay         
    Pass: 1
    Read option: roundrobin (default)
    Write option: parallel (default)
    Size: 30721344 blocks (14 GB)

d1: Submirror of d0
    State: Needs maintenance 
    Invoke: metareplace d0 c0t0d0s0 <new device>
    Size: 30721344 blocks (14 GB)
    Stripe 0:
        Device     Start Block  Dbase        State Reloc Hot Spare
        c0t0d0s0          0     No     Maintenance   Yes 


d2: Submirror of d0
    State: Okay         
    Size: 30721344 blocks (14 GB)
    Stripe 0:
        Device     Start Block  Dbase        State Reloc Hot Spare
        c0t8d0s0          0     No            Okay   Yes 


Device Relocation Information:
Device   Reloc  Device ID
c0t8d0   Yes    id1,sd at SSEAGATE_ST373307LSUN72G_3HZ68NAX000074173FY8
c0t0d0   Yes    id1,sd at SSEAGATE_ST373307LSUN72G_3HZ6NA8500007424CC3V


David Martini
LLNL
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From kristianto.setiawan at sun.co.id  Mon Sep 20 21:57:59 2004
From: kristianto.setiawan at sun.co.id (kristianto.setiawan at sun.co.id)
Date: Tue, 21 Sep 2004 01:57:59 -0000
Subject: SUMMARY : Sum in Script
Message-ID: <957b1ffa7269.414fec39@sun.co.id>

Thanks everyone who answered.

The simplest answered from : 

1. No Spam <nouce at mighty.co.za>  
cat /data/vxdg-g_free=sapdg1.out|grep -v DISK|awk '{tot+=$5};END{ print tot}'

2. Chunk Lin  (chunklin at lucent.com)
cat  /data/vxdg-g_free=sapdg1.out | nawk 'BEGIN{sum = 0}; FNR > 1 { sum += $5 }; END{ print sum}'

3. Sebastien Dubigne (sebastien.daubigne at atosorigin.com) count in Mbytes
vxdg -g <DG> free | awk '/[0-9]/{s+=$5}END{print s/2048}'

Other send me script, but many didn't works.

Rgds,
Kristianto

----- Original Message -----
From: kristianto.setiawan at sun.co.id
Date: Saturday, September 18, 2004 5:38 am
Subject: Sum in Script

> Hi script  expert :
> I have a data like this :
> # cat  /data/vxdg-g_free=sapdg1.out
> DISK               DEVICE       TAG          OFFSET    LENGTH    FLAGS
> sapdg101     c2t0d0s2     c2t0d0       165       10011     -
> sapdg101     c2t0d0s2     c2t0d0       16787392  3008      -
> sapdg101     c2t0d0s2     c2t0d0       50344832  6016      -
> sapdg101     c2t0d0s2     c2t0d0       58739456  6592      -
> sapdg101     c2t0d0s2     c2t0d0       58746147  10077     -
> sapdg101     c2t0d0s2     c2t0d0       67144832  6592      -
> sapdg101     c2t0d0s2     c2t0d0       67151523  10077     -
> sapdg101     c2t0d0s2     c2t0d0       143328768 192       -
> sapdg102     c2t1d0s2     c2t1d0       16777216  3008      -
> sapdg102     c2t1d0s2     c2t1d0       50334656  6016      -
> sapdg102     c2t1d0s2     c2t1d0       50340969  9879      -
> sapdg102     c2t1d0s2     c2t1d0       58739456  6592      -
> sapdg102     c2t1d0s2     c2t1d0       67134656  6592      -
> sapdg102     c2t1d0s2     c2t1d0       143308416 192       -
> sapdg102     c2t1d0s2     c2t1d0       143309235 19725     -
> 
> How to create the shell script  to sum only the length column (5th) 
> to get total free space disk from that data above ?
> 
> Rgds,
> Kristianto
> _______________________________________________
> 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 Pit-Ong.Ong.Goh at reuters.com  Tue Sep 21 03:15:55 2004
From: Pit-Ong.Ong.Goh at reuters.com (Pit-Ong.Ong.Goh at reuters.com)
Date: Tue, 21 Sep 2004 07:15:55 -0000
Subject: SUMMARY: secure ways of disposing a hard disk - command needed
 to erase data
Message-ID: <T6c2c7787cac407b710bcc@dtcseuvig4.dtc.lon.ime.reuters.com>

Thank you all,  lots of great stuff I received.


Rgds
Goh

================================================

In the past I've gone for physical damage to the drive.
i.e. you take the drive out the back with a sledge hammer.

I've not heard of anyone ever recovering data from a drive with
bent/broken platters, and it's good stress relief too!

If you want the drive to be usable again afterwards you could do one of
the destructive write tests in /usr/sbin/format.

Steve
"Bennett, Steve" <s.bennett at lancaster.ac.uk>
09/20/2004 11:30 PM


=====

Here is a link that I think will be of more
use: http://slashdot.org/askslashdot/01/02/21/1752256.shtml



Wade Lich


=======

Hi Goh,


The best way we do it is by wiping the disk out using
strong magnets, but if thats not an option for you,
you can format the disk, and fill it up using DD

dd of=/dev/rdsk/c0t4d0s0 if=/dev/zero

you can do this for each partition.. this shoudl wipe out
most of the data.


If you have any questions, please let me know.

Thanks!

...Chintu


=======

I dd /dev/zero onto each sector seven times and then throw the disk away.

dd if=/dev/zero of=/dev/rdsk/cXtXdXs2 bs=1024

Between each run I need to put a disk label back onto the disk, but that
is done very quickly by "format".



"Christopher L. Barnard"
09/20/2004 09:46 PM

=======


I found this information about a year ago while searching for similar info.
I don't have the original posters name so I'm unable to attribute credit
where due :

Believe it or not, there is an actual standard for "drive cleaning".
Repartition, newfs or SCSI G0 format is not a drive cleaning.

To clear the data out of a drive you need to
a) write all 1s
b) write all 0s
c) write 010101010... across the whole drive
d) create new filesystems (probably needs a new partition first,
and why not do a G0 format while you are at it).

This is somewhat tedious, as each a), b) and c) kills the partition
table requiring a new partition.

Any other method makes data recovery (theft?) tedious, but possible.

Do not perform a G0 format first -- remapping bad blocks might result
in data be hidden but not removed.

b) above can be done with
             dd if=/dev/zero bs=1024k of=<raw disk>

a) and c) require either either special programs, or special
(non-existant) devices.  What you want is something like

             dd if=/dev/0xAAAAAAAA of=<raw disk>
             and
             dd if=/dev/0xFFFFFFFF of=<raw disk>

but you will probably have to write a few simple programs so you can

        dev_0xAAAAAAAA | dd of=<raw disk>
             and
             dev_0xFFFFFFFF | dd of=<raw disk>

Mike Galvez

=======

secure sites for the us goverment use an electic drill.

other than physical destruction, there is no way to totally destroy the
data.
us government standards work very well for most businesses.

most methods uses overwriting the entire with random data serveral
times.
(different random data each time.)
ending with an all 0 or 1 finishes the process.
this works well for most businesses.
but, a determined foe with lots of money and time can reconstruct the
data.
(track drifts, etc. makes this possible.)

the electric drill works best.


=====

Hello there,

Overwrite the entire disk 7 times.
Just dump general data on it to overwrite it.  Make sure you write to all
partitions.  I usually just:

1) mount the partitions of the expiring disk
2) delete everything on those partitions
3) write to it.   you could truss a bloated process.  `truss -p $PID >
/scrapdisk/file.foo`.
(don't use mkfile because size is only noted, but  disk blocks  are  not
allocated  until  data is written to that file)
4) once it's full, delete it and repeat #3 about 6 more times

A better way:
*) De-gaussing!  demagnetize the disk.  get a huge high powered magnet and
'zap' the disk 2 times.  it'll make it unusable.  caveat: don't do the
anywhere near your wallet (credit cards, etc), your office access badge or
anything else that's magnetic.  an it's a pain because the proper way is to
disassemble the disk because the platters have a small amount of sheilding.

Regards
Daryl.Mitchell at us.o-i.com
09/20/2004 08:39 PM

======

Newfs on the disks and this will distroy all data on the disk.
This is native to solaris.
Thanks
Vinnie German
NSI

=======

I find that a T6 torx bit on a screwdriver is compatible with a wide range
of OSs for ensuring that the disks are no longer usable, readable.  Just
pop the casings open, remove the platters and attack them with either
strong magnets (two included FREE in each disassembled hard disk !) or
scratch liberally.

Takes a couple of minutes per disk, once you've had a little practice.


the hatter

=======

afaik, format - analyze - verify will destroy the data
Otherwise you can use dd if=/dev/zero of=/dev/disk but this is not
100% sure (if you have a /dev/random it would be better to use that 2
or 3 times).

I normally just use /dev/zero on the first blocks but my security
needs are minimal. (enough for casual scrubbing)

Ronny Martin

=======

just format the disk with a new partition that encompasses the whole disk, then newfs it.
do you back it up?  think about getting rid of the backups as well if necessary.

Fonseca Mario

======

How secure do you want to be?  And how much time do you have?

while :
do
             dd if=/dev/zero of=/dev/dsk/c?t?d?s2
             dd if=/dev/random of=/dev/dsk/c?t?d?s2
done

Will wipe the disk reasonably well, as long as you leave it running long enough.  The UK
government used to state that a write of 0, then 1, then random, should be done 99
times before the disk is "safe", but this may be overkill for anything short of the plans of
how to build a nuclear missile.  Not sure if this is still their recommendation

The use of random rather than always 0 or 1 is key though, so although this is a lot slower
than using just all zero, then all one, then repeat, it's also a lot more secure.

Graham


=======

Our risk management people accept five passes of format's "write" test with
random data. This generally takes all night.

You can't run this on the boot disk! Normally, I place the disks to scrub in
an old E250 I have here and run format. When it has finished, I install the
disks back in the original system and then jumpstart it to put a basic OS
back on. The 250 can handle four or five disks at once.

For workstations, I boot them off the network into single user mode (ok boot
net -s) and then run format.

The format options:

format -> select disk -> analyze -> set up (defaults, except 5 passes, and
random patterns.),
then analyze -> write.

Sometimes you need to run format -> type to reset the disk type after this,
so it is probably a good idea to run format -> current to get that
information before you start the write sweep.


-- Russell Page.


=======

dd if=/dev/random of=/dev/rdsk/cXtXdxs2 bs=128k

"Smith, Kevin"

======


burn the hard disks in an incinerator
is the only way if you have military or HPPIA-controlled data on the
disks.

Else do a search on www.sun.com/blueprints for a "format" script
which will copy 1 and zeros 4 times across the entire disk.

JV711

=====


      09/20/2004 11:56 AM

You use either use dd if=/dev/null of=/dev/<your disk>

or
In the format command there is an option to destroy the data.

Hope these helps.


=====

several iterations of:

dd if=/dev/zero of=/dev/rdsk/c0t0d0s2

should do the trick - it writes all zero's to the entire disk.  I would
do it more than once just to be safe.

-Jon


=======


Screwdriver and a file/metal bruch/sander...  Pull the drive/platters
apart and give them all a good scrubbing.

Good luck
Forrest


=========  Original question =================

From: sunmanagers-bounces at sunmanagers.org
[mailto:sunmanagers-bounces at sunmanagers.org] On Behalf Of
Pit-Ong.Ong.Goh at reuters.com
Sent: Sunday, September 19, 2004 10:45 PM
To: sunmanagers at sunmanagers.org
Subject: secure ways of disposing a hard disk - command needed to erase
data

Hi,

Perhaps this has been asked before :
We need to securely scrap/dispose hard disks.

What's the various ways of doing it aside from using 3rd party tools.

We want something that's native to Solaris 2.6 to Solaris 8, so secure
that you can recover the data anymore.

Hopefully it's something simple/easy.


Thanks
Goh




--------------------------------------------------------------- -
        Visit our Internet site at http://www.reuters.com

Get closer to the financial markets with Reuters Messaging - for more
information and to register, visit http://www.reuters.com/messaging

Any views expressed in this message are those of  the  individual
sender,  except  where  the sender specifically states them to be
the views of Reuters Ltd.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From suporte at os.com.br  Tue Sep 21 08:17:27 2004
From: suporte at os.com.br (OS)
Date: Tue, 21 Sep 2004 12:17:27 -0000
Subject: Summary: Solstice Backup
Message-ID: <KEEHIBNKIEHMGLKICLKDCEINCBAA.suporte@os.com.br>

Hi,

 Thanks a lot for everybody who help me.

  One way to control the size of indexes is using the gui interface nwadmin:
Clients -- Indexes -- Remove --- Reclaim space.

  By command line:

  nsrck -P | -d | -m -R | Volumename
  nsrck -C clientname

  Thanks,

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


From martini at mrpeabody.llnl.gov  Tue Sep 21 18:03:11 2004
From: martini at mrpeabody.llnl.gov (Dave Martini 1)
Date: Tue, 21 Sep 2004 22:03:11 -0000
Subject: SUMMARY: Trouble connecting to a 3310 Array Using ssconsole
Message-ID: <200409212201.i8LM10909736@raider.llnl.gov>

Well my brain deffiniely isn't working well today.
I was putting in the IP address of the ARRAY instead of the HOST under
server setup in the ssconsole GUI which is deffinitely why it didn't work.

I went through the configuration steps and now I can also telnet into the box.

Here's my original question.

I have a V250 with Solaris 9 on it.
I installed the CLI packages
 1  SUNWsccli     Sun StorEdge(tm) 3000 Family CLI
                   (sparc) 1.5.0,REV=2004.02.03.12.27
  2  SUNWscsa      Sun StorEdge(tm) Diagnostic Reporter daemon
                   (sparc) 1.5.0,REV=2004.02.03.13.26
  3  SUNWscsd      Sun StorEdge(tm) Configuration Service Agent
                   (sparc) 1.5.0,REV=2004.02.02.13.24
  4  SUNWscsu      Sun StorEdge(tm) Configuration Service Console
                   (sparc) 1.5.0,REV=2004.02.03.13.25
  5  SUNWscui      Sun StorEdge(tm) Diagnostic Reporter console
                   (sparc) 1.5.0,REV=2004.02.03.13.26
and The daemons are running

 root   235     1  0 09:34:17 ?        0:00 /usr/sbin/ssmon
    root   231     1  0 09:34:16 ?        0:00 /usr/sbin/ssserver

I can connect to the Array via the tip command no problem. I've configured the 3310 using
tip and can see the array no problem. I added an IP address to the 3310 and can ping the 
Array from any host.

Trouble is when I execute ssconsole and add the array I get the "discovering" message
and it never connects to the array.

Are there any tricks to get ssconsole to connect to the 3310?

I was told by the SUN rep that I  have to be able to telnet into the array before ssconsole 
can connect. Is this correct? When I try to telnet into the array it just says trying to 
connect but never does. I didn't notice any references to telnet in the CLI manual.

Thanks much!
David Martini
LLNL
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From sun at percipia.com  Tue Sep 21 19:18:05 2004
From: sun at percipia.com (Sundaram Ramasamy)
Date: Tue, 21 Sep 2004 23:18:05 -0000
Subject: SUMMARY: disk not responding to selection
Message-ID: <1376.68.75.19.214.1095808649.squirrel@68.75.19.214>

Hi all,

My Thanks to every respond for this problem. Hard disk failed we replaced
the hard disk.

==============

Looking first at the DiskSuite messages ("State: Needs maintenance") this
typically means that there is something wrong with the DiskSuite LUN, or a
component of that LUN.  Most often, it is that one or more physical disks
have failed.

Looking then at the /var/adm/messages output you have supplied, this is
confirmed - the disk drive (/dev/dsk/c0t0d0s<anything>) has failed, and
needs to be replaced.

Once you have physically replaced the disk with a new one of the same
type, you can use the following command to format the drive the same as
its mirror drive:

  # prtvtoc /dev/rdsk/c0t1d0s0 | fmthard -s - /dev/rdsk/c0t0d0s0

(this just sets up the partition table on the new disk, which
I am assuming will directly replace the old, failed c0t0d0
in the same slot, with the same partitioning as its mirror pair disk).

Once that is done (and you can double-check it in "format"), you should be
able to issue the series of "metareplace" commands needed to restore the
mirroring state of this set of LUNs, like so:

  metareplace -e d10 c0t0d0s0
  metareplace -e d20 c0t0d0s1
  metareplace -e d30 c0t0d0s3
  metareplace -e d40 c0t0d0s4

(the '-e' option just tells DiskSuite to replace the disk "in place" -- or
just replace a bad disk with a good one in the exact same location).
=============

 There is problem in one of the disk and it has gone
faulty. Check the disk name/position by format command
for this disk /pci at 1f,4000/scsi at 3/sd at 0,0.
You have to replace the disk and then you have to
repair the metadisks.

-- 
-Sundaram Ramasamy
SCSecA,SCNA,SCSA,SCWCD,SCJP
Storage & Sun Solaris Administrator
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From chipman at ecopiabio.com  Wed Sep 22 10:20:49 2004
From: chipman at ecopiabio.com (Tim Chipman)
Date: Wed, 22 Sep 2004 14:20:49 -0000
Subject: SUMMARY:  force-mount a corrupt UFS FS possible w/o fsck
 completion ?
Message-ID: <41518A20.1040803@ecopiabio.com>

Many thanks (!!) to all who responded (no particular order: David Deaves, Alan Pae, NoUce "E", Terry Moore, Darren Dunham, Angelo Bonfieto Jr., Charles Rawls).  The concensus was loud and clear, mount the (!&$#@@) thing with read-only mount option, and things should work more smoothly:

mount -o ro /dev/dsk/bad-device /mount-point

Failing that, if the device is still giving excessive grief, "DD" might be used to rip raw data directly from the device path and dump it to another local safe storage area. [Fortunately I didn't have to resort to this]

Failing that, if the data missing is of great enough value, 3rd party services are happy to accept your $$ to make an attempt at recovery :-)

In my case, read-only mount was adequate to rescue the *very few* (literally - 5 files total size of < 1 meg on a 1.4Tb raid5 array) files that had been modified since the last good backup done the night before.  Certainly at times like this, one gains a greater appreciation for reliable backup systems :-)


Thanks again to all,

---Tim Chipman



===ORIGINAL QUESTION===
A fun "hypothetical question" (apparently un-answered in google/list
archives)

Is it possible to force-mount a known corrupt volume (ie, it cannot finish
FSCK and if mount is attempted it gives "fatal read errors") -- **purely**
for the purpose of attempting to recover *some* small portion of data on the
volume .. ?

context,

-failed blocks on too many disks in a raid5 array. One disk in particular
failed totally.

-the thing is in bad shape. Rebuild of array will be 4-5 days based on the
first 24 hours of performance observed. Uncertain it will actually finish
successfully.

-recent backup renders this a not-disaster.

-small amount of data .. not backed up but would be nice to recover if
possible

[Filesystems are UFS/Logging Solaris8 ]


Any thoughts ?


Many thanks!


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


From daniel.denes at bankgesellschaft.de  Wed Sep 22 10:31:50 2004
From: daniel.denes at bankgesellschaft.de (Denes, Daniel (BG-IT 51))
Date: Wed, 22 Sep 2004 14:31:50 -0000
Subject: SUMMARY: Replacing A1000 batteries...online
Message-ID: <3641E2ABFD3EDD48B09F33361B0B8ADADFD68E@TZEX020.bgb.de>

I asked for best practises and the supported method to replace A1000 batteries
(with focus on wether or not its wise to do it online). I received a number of
very helpful replies which i will summarize below. First - apologies for the
very late summary, unfortunately our vendor took >2months to provide us with
the 4 neccessary batteries 8^(

===== Responses (email addresses removed for spam prevention) =====
Albert Musin wrote:
Hi,

I did it online as written in RM instruction (recovery guru).
It was OK.
-------------------------------------------------
ponder wrote:
While sun still claims it is impossible, we have changed about 20 batteries
without turning the cabinets off in half a day. Don't forget to reset the
battery age (can't remember command at the moment, check osa docs). And
ofcourse this procedure is completely unsupported by sun...
-------------------------------------------------
Sudesh Shetty told me:
Hi Daniel,
          You can replace the battery online as i had done it before on an
A1000 connected to a production server.
-------------------------------------------------
Valuable insight from Sun came from Tony Walsh:
It has never been a Sun engineering supported procedure to "hot swap"
batteries
in an A1000 array (rumoured to be related to litigation related..). The Sun
statement of support states that a battery exchange should be made with all
power
removed from the array. I have found that this policy can often lead to much
more
drastic failures if the array has other issues that are not apparent (or not
checked) before power is removed. It also gives disk drives a chance to cool
down
(even for a short period) and makes it much more likely that the drive may
not
spin up when power is reapplied. Not to be deterred by any of this, the
approach
I usually use is to give the customer the choice and advise that data lose is
a
possiblity, and then use the following process:-
- Determine if the host can be shut down to the OK> prompt. If it can, this
will
stop all IO to the array and prevent data loss. At this point I just "warm
swap"
the battery and make sure it begins charging.
- If the host must remain active the next level of data safety is to unmount
all
the file systems on the host(s), wait 10-15 minutes to ensure the cache is
flushed,
then exchange the battery and make sure it starts charging.
- If all filesystems must remain mounted, I then turn off all "write through"
cacheing using the RM6 command line or gui, wait at least 10-15 minutes to
ensure
the cache is flushed, and then change batteries.

I have never yet had a data loss event or any other electrical misadventure,
but it
must be said again, the risk is all in your hands as the batteries are not
designed
as a "hot swap" item.

In your case, as the batteries are already well beyond their "use-by" date,
the cache
"write through" has already been turned off, so I would just pull each battery
in
turn and replace it with a new one (replace the first battery before removing
the next
one). The only caution I would give is, make the change during a low activity
period
just in case there is an electrical misadventure.

One other warning is to make sure the battery has been recharged within the
last 6
months or the recharge operation in the array may take too long and then cause
the
"new" battery to be marked bad again.
-------------------------------------------------
Gene Beaird has a different practise:
I have been working with A1000s since about 2000, and have never heard of
doing the battery change online.  We have always scheduled an outage,
powered the system off and swapped the batteries.  Should have maybe tried
it to one of our standby systems when I had the chance, but never did.  I
guess the potential power surge when you pull the battery out can fry some
of the electronics on the system.
-------------------------------------------------
Tim Chapman pointed out, that
For what it is worth, a semi-recent summary to the list reiterates the
requirement, "turn off, swap battery, power on, issue 'battery age
reset' command" ...

http://www.sunmanagers.org/pipermail/summaries/2002-July/002014.html

maybe others will suggest otherwise ?
-------------------------------------------------
Tod Sandman sent me this very helpful recipe:
I always replace live by first turning off caching.  I too heard
rumors etc. and so recently verified my procedure with our Sun
onsite guy.  For what it's worth, he says yep, that's the way to
do it.  Let me know if you hear otherwise.

Here are my typical steps:

## Define these 2:
device=c1t5d0s0; luns=0,1,2,3,4

## To check battery:
  raidutil -c ${device} -B
## To replace battery:
  ## Turn off caching if it is still enabled:
    raidutil -c ${device} -w off ${luns}
  ## Replace the battery.
  ## Reset battery age:
    raidutil -c ${device} -R
  ## Re-enable cache after replacing battery:
    raidutil -c ${device} -w on ${luns}
-------------------------------------------------
Dirk Boenning summarized like this:
Hello,

Sun: You have to shutdown the whole system.

Experience: Be sure to disable the cache. Change battery and enable cache
again. Done couple of times without any problem.
-------------------------------------------------
Gene Siepka knew:
You can definitely do it online. Our local Sun Field Engineer says its like a
urban legend that you must power off...
I've done it on a A1000 just 2 months ago. No problem.
-------------------------------------------------
My original mail:
Hello List,

we have a couple of A1000 Arrays with 8 and 12 disks. All of their batteries
have exceeded their lifetime long ago and are to be considered disfunctional.
As we hope to gain performance from a working cache we want to replace them;
however as they are attached to HA-systems we would clearly prefer to do that
"online", i.e. without shutting them down. I recall it used to be a supported
procedure to replace the batteries like that, but i heard rumors that Sun has
backed up on that. Does anyoen have insights, experiences, rumours (s)he could
share on that topic?
-------------------------------------------------
What i did:
- used rm6 (shame on me for using a GUI) to turn cache off (of course it was
disabled for years already, the oldest of the replaced batteries dated back to
1998)
- got myself a console window to see kernel output, also had a look on
messages
- pulled old battery online (but, as recommended, at a time where it would
hurt the minimal number of users if something bad happened). Pulling it out
proved to be somewhat difficult for two of them because the adhesive labels on
top suffered and rolled themselves up as i dragged the cannister out, i had to
use some force to get them out the last few centimeters.
- slid new one in
- watch console and messages - eventually 3 messages per battery would show
up, looking like this:
raid: [ID 702911 user.error]
Sense=700006000000009800000000A0000000000000000000000000000000000000000000800
000082C000000000000000000000000000B053154383032303130303720202020202003010300
00010000000000000000000000000000000000000000000000000005000000000000000000000
0000000000000000000000000000000000000000000000009A412A93039313730342F30393339
353400000000000000
no need to panic though its just this one message.
- reset battery age as suggested by Tod by doing raidutil -c ${device} -R
- turned caching on again (with rm6 again, who told me right away that cache
was temporarily disabled again while the batteries were loaded).
- checked cache status 20 min later, it was on and working
- tested performance with iozone, performance improvement was like factor 2.5

Thanks everyone for the insights.

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


From ghenry at cmi.univ-mrs.fr  Thu Sep 23 10:23:48 2004
From: ghenry at cmi.univ-mrs.fr (=?ISO-8859-1?Q?G=E9rard_Henry?=)
Date: Thu, 23 Sep 2004 14:23:48 -0000
Subject: [SUMMARY]: can't send break in tip
In-Reply-To: <4152C77E.7070404@cmi.univ-mrs.fr>
References: <4152C77E.7070404@cmi.univ-mrs.fr>
Message-ID: <4152DC09.9090504@cmi.univ-mrs.fr>

many thanks for fast replies from a lot of people.
The answer seems simple:
for example, h. kreft said:
"Because ~# is interpreted by the SSH you used to connect to said blade
100. If you want to send tip the sequence '~#' you may either change the
escape character of your ssh client by specifying the -e parameter"

i forgot to say i'm using ssh!
another suggestion was to do:
"Try ~~# after a newline. You have to escape ~ so tip can see ~ as its
escape character."
it can't works for me, i don't know why.

Finally, i just did:
~s to change value of escape character, here i use $, and then type $$#
and it works fine.

thanks to:
Roland.Merk
Raymond.Plassart
bryan pepin
remy.zandwijk
rene occelli
chris pinnock
helmut kreft


> i've a machine administered with a serial cable beetween port A and port 
> B of a sun blade 100.
> On sun blade 100, in a terminal, i use tip -9600 /dev/cua/b to connect.
> I want to send a break to the remote machine, with ~#, but it don't work:
> ~#
> The following connections are open:
>   #0 client-session (t4 r0 i1/0 o16/0 fd 6/7)
> 
> and if i do:
> ~?
> Supported escape sequences:
> ~.  - terminate connection
> ~^Z - suspend ssh
> ~#  - list forwarded connections
> ~&  - background ssh (when waiting for connections to terminate)
> ~?  - this message
> ~~  - send the escape character by typing it twice
> (Note that escapes are only recognized immediately after newline.)
> 
> i see that ~# is not for send a break, why?
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From edwin.sadeh at rbc.com  Thu Sep 23 12:02:55 2004
From: edwin.sadeh at rbc.com (edwin.sadeh at rbc.com)
Date: Thu, 23 Sep 2004 16:02:55 -0000
Subject: Summary -- locking accounts after 3 unsuccessful login attempts
Message-ID: <C854197844209A4C8DEE9BB449FAD5090140BC9C@sew01038.maple.fg.rbc.com>

Hello,

I was reading up on this issue and came up to this page. I was wondering If
you could send me the reply from Darren Moffett of sun to this problem.

Regards
Sincerely

Edwin Sadeh
Information Security
Tel (416) 348-5441
edwin.sadeh at rbc.com


------------------------------------------------------------
This e-mail may be privileged and/or confidential, and the sender does not waive any related rights and obligations. Any distribution, use or copying of this e-mail or the information it contains by other than an intended recipient is unauthorized. If you received this e-mail in error, please advise me (by return e-mail or otherwise) immediately. 

Ce courrier ilectronique est confidentiel et protigi. L'expiditeur ne renonce pas aux droits et obligations qui s'y rapportent. Toute diffusion, utilisation ou copie de ce message ou des renseignements qu'il contient par une personne autre que le (les) destinataire(s) disigni(s) est interdite. Si vous recevez ce courrier ilectronique par erreur, veuillez m'en aviser immidiatement, par retour de courrier ilectronique ou par un autre moyen.

============================================================


From dm at hejoe.de  Fri Sep 24 04:30:51 2004
From: dm at hejoe.de (Dirk Maass)
Date: Fri, 24 Sep 2004 08:30:51 -0000
Subject: [SUMMARY]: How can we make a multihomed server replying on the
 same interface
In-Reply-To: <4151911D.4010106@hejoe.de>
References: <4151911D.4010106@hejoe.de>
Message-ID: <4153DB1A.1030109@hejoe.de>

Dirk Maass wrote:
 > Is there a way to make that the server sends his answer packet out on
 > that interface where the request arrived?
 >

Thanks to all of you who replied to my question,
especially to Adam Levin who gave me more insight.

Although I haven't solved my case yet, I want to summarize because all 
answers were the same: NO.

Casper Dik wrote:
> No; the information is lost as soon as the packet arrives so the
> system cannot know where to send it back.

Darren Dunham wrote:
> #2, There's not always a concept of "the answer packet".  Although it
>     makes a lot of sense for NFS, there's nothing that says an input
>     packet has to cause an output packet.  So output can be based on
>     several inputs, and then there's no easy way to align them.
> ...
> My concern is that if you're
> doing UDP, there's no connection, so the sender IP address could be
> different.

Casper and Darren also suggested using NFS over TCP, "but that would 
only address NFS and not the rest" (Casper)

A static host route as Drew Skinner suggested would not solve my sroblem 
as there are many Firewalls not only one. I tried to set up a static 
network route in the server to the network where the Firewalls are 
located. But that didn't help. With tcpdump I still saw server-packets 
from different IP addresses.

Adam helped me to understand that the server would send out the 
UDP-packet on that interface wich it believes it's closest to the 
client. So the solution may also lay in a router/network config and may 
probably result in a simpler network setup.
I'll have to think about it and to discuss the problem deeper with the 
networkers because I have no access to these components here.

Thank you all, again.
Dirk
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From ahoesch at smartsoft.de  Fri Sep 24 15:41:17 2004
From: ahoesch at smartsoft.de (=?ISO-8859-1?Q?Andreas_H=F6schler?=)
Date: Fri, 24 Sep 2004 19:41:17 -0000
Subject: Summary: DNS problem on Solaris 9
In-Reply-To: <957C8498-0E60-11D9-9923-003065CCA582@smartsoft.de>
Message-ID: <A48F18C6-0E61-11D9-9923-003065CCA582@smartsoft.de>

Hello,

I rechecked the config for the 10th time and finally discovered a  
difference between my working server and the one I have here. I had  
<zone "0.1.168.192.in-addr.arpa" > in /etc/named.conf. This seems to  
have been the show-stopper. I finally managed to get it working. Here  
are the relevant pieces.

$TTL 3h
1.168.192.in-addr.arpa. IN      SOA  smartsoft-hl.de.  
root.smartsoft-hl.de. (
                                 2004092410      ; Serial
                                 86400           ; Refresh
                                 7200            ; Retry
                                 604800          ; Expire
                                 86400           ; Minimum
)

1.168.192.in-addr.arpa. IN      NS      smartsoft-hl.de.
1       IN      PTR     merkur.smartsoft-hl.de.

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

zone "1.168.192.in-addr.arpa"
{
         type master;
         file "smartsoft-hl.de.rev";
};

Thanks to all!

Regards,

   Andreas

> I am back at my Solaris 9 server and still had no luck getting DNS to  
> work. I have a very similary configured Solaris 8 box that works. If I  
> execute "nslookup -d2" there I get
>
> *********************************************************************** 
> ************
> bash-2.05# nslookup -d2
> ;; res_nmkquery(QUERY, 1.1.168.192.in-addr.arpa, IN, PTR)
> ------------
> SendRequest(), len 42
>     HEADER:
>         opcode = QUERY, id = 43558, rcode = NOERROR
>         header flags:  query, want recursion
>         questions = 1,  answers = 0,  authority records = 0,   
> additional = 0
>
>     QUESTIONS:
>         1.1.168.192.in-addr.arpa, type = PTR, class = IN
>
> ------------
> ------------
> Got answer (115 bytes):
>     HEADER:
>         opcode = QUERY, id = 43558, rcode = NOERROR
>         header flags:  response, auth. answer, want recursion,  
> recursion avail.
>         questions = 1,  answers = 1,  authority records = 1,   
> additional = 0
>
>     QUESTIONS:
>         1.1.168.192.in-addr.arpa, type = PTR, class = IN
>     ANSWERS:
>     ->  1.1.168.192.in-addr.arpa
>         type = PTR, class = IN, dlen = 25
>         name = ns.advanced-science.com
>         ttl = 10800 (3H)
>     AUTHORITY RECORDS:
>     ->  1.168.192.in-addr.arpa
>         type = NS, class = IN, dlen = 2
>         nameserver = advanced-science.com
>         ttl = 10800 (3H)
>
> ------------
> Default Server:  ns.advanced-science.com
> Address:  192.168.1.1
> *********************************************************************** 
> ************
>
> If I do the same on my Solaris 9 box I instead get
>
> *********************************************************************** 
> ************
> bash-2.05# nslookup -d2
> ;; res_nmkquery(QUERY, 1.1.168.192.in-addr.arpa, IN, PTR)
> ------------
> SendRequest(), len 42
>     HEADER:
>         opcode = QUERY, id = 49175, rcode = NOERROR
>         header flags:  query, want recursion
>         questions = 1,  answers = 0,  authority records = 0,   
> additional = 0
>
>     QUESTIONS:
>         1.1.168.192.in-addr.arpa, type = PTR, class = IN
>
> ------------
> ------------
> Got answer (139 bytes):
>     HEADER:
>         opcode = QUERY, id = 49175, rcode = NXDOMAIN
>         header flags:  response, auth. answer, want recursion,  
> recursion avail.
>         questions = 1,  answers = 0,  authority records = 1,   
> additional = 0
>
>     QUESTIONS:
>         1.1.168.192.in-addr.arpa, type = PTR, class = IN
>     AUTHORITY RECORDS:
>     ->  168.192.in-addr.arpa
>         type = SOA, class = IN, dlen = 65
>         ttl = 10800 (3H)
>         origin = prisoner.iana.org
>         mail addr = hostmaster.root-servers.org
>         serial = 2002040800
>         refresh = 1800 (30M)
>         retry   = 900 (15M)
>         expire  = 604800 (1W)
>         minimum ttl = 604800 (1W)
>
> ------------
> *** Can't find server name for address 192.168.1.1: Non-existent  
> host/domain
> *** Default servers are not available
>
> It looks like someone else feels responsible for 192.168.  I am not  
> sure what this means. Here is my config.
>
> /var/named/smartsoft-hl.de.rev
> $TTL 3h
> 0.1.168.192.in-addr.arpa.       IN      SOA  merkur.smartsoft-hl.de.  
> root.smartsoft-hl.de. (
>                                 2004092410      ; Serial
>                                 86400           ; Refresh
>                                 7200            ; Retry
>                                 604800          ; Expire
>                                 86400           ; Minimum
> )
>
> 0.1.168.192.in-addr.arpa.       IN      NS      merkur.smartsoft-hl.de.
> 1       IN      PTR     merkur.smartsofzone "0.1.168.192.in-addr.arpa"
>
> /etc/named.conf
> ...
> zone "1.168.192.in-addr.arpa"
> {
>         type master;
>         file "smartsoft-hl.de.rev";
>         allow-transfer {
>                 local_network;
>         };
> };
>  ...
>
> /etc/resolv.conf
> domain smartsoft-hl.de
> search smartsoft-hl.de
> nameserver 192.168.1.1
>
> Any ideas?
>
> Thanks,
>
>   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 ifetch at du.edu  Fri Sep 24 17:43:42 2004
From: ifetch at du.edu (Ivan Fetch)
Date: Fri, 24 Sep 2004 21:43:42 -0000
Subject: Summary: Install Solaris 9 X86 Via Serial Port?
Message-ID: <20040924153839.T58935@tnetnzry.pnve.qh.rqh>

Hello,

    Thanks to the following for their responses:

Casper Dik <casper at holland.sun.com>
Alex Stade <alex at trdlnk.com>
Peter Ondruska <peter at ondruska.com>
Simon Burr <simes at bpfh.net>


    The consensus was to use ttya instead of com1, which worked perfectly 
(I really should have tried that, despite docs.sun.com and other searches 
thus far stating that ttya does not work but com1 does).


(my initial question)

Date: Fri, 24 Sep 2004 09:53:49 -0600 (MDT)
From: Ivan Fetch <ifetch at du.edu>
To: sunmanagers at sunmanagers.org
Subject: Install Solaris 9 X86 Via Serial Port?

hello,

     i'm attempting to install Solaris 9 X86 via the serial port by adding 
the following lines to /solaris/bootenv.rc on the boot floppy (created 
from the image file on the Software CD #2):
setprop output-device com1
setprop input-device com1


     (I made these changes to the bootenv.rc file on  a Linux box)

     On boot it says:
Solaris boot sector, version 1

Solaris mdb exec load error


     Does anyone have experience doing an install via serial port  like 
this, that can offer any suggestions?

Thanks in advance for your help -- ivan Fetch.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From mrb at bmyster.com  Wed Sep 29 08:29:37 2004
From: mrb at bmyster.com (Brent Bailey)
Date: Wed, 29 Sep 2004 12:29:37 -0000
Subject: SUMMARY: DiskSuite on Solaris 9
Message-ID: <50042.65.175.128.10.1096460760.squirrel@new.host.name>

My question was does solaris 9 use Solstice DiskSuite or is there
something else that Sol 9 uses.


ANSWER: DiskSuite has been changed to Sun Logical Volume Manager (LVM) and
it comes bundled with Solaris 9.

Thank you everyone for your replies.
-- 
Brent Bailey
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From DTeklu at amlaw.com  Wed Sep 29 10:42:13 2004
From: DTeklu at amlaw.com (Daniel Teklu)
Date: Wed, 29 Sep 2004 14:42:13 -0000
Subject: SUMMARY..RE: unrecognized option `-KPIC'
Message-ID: <2F24F38510C7E144AFD857EE277BB0412A27F1@alm-mail-1.nyc.amlaw.corp>

Thanks to all who replied. Sanjiv K. Bhatia   got it.

In my make file I changed
OPTIMIZE = -xO3 -xdepend

to 

OPTIMIZE = O3

and that did the trick.

-Daniel 




> ----------
> From: 	sunmanagers-bounces at sunmanagers.org on behalf of Daniel
> Teklu
> Sent: 	Tuesday, September 28, 2004 3:44 PM
> To: 	sunmanagers at sunmanagers.org
> Subject: 	cc: unrecognized option `-KPIC'
> 
> I am trying to install a perl module on solaris 9. I keep getting this
> error:
> 
> cc: unrecognized option `-KPIC'
> 
> I installed gcc from sunfreeware.com and I removed the original 'cc' and
> 'make' and put the new /usr/ccs/bin/make and /usr/local/bin/gcc in the
> path.
> But still getting the error.
> 
> Any ideas?
> 
> Thanks in advance.
> 
> -Daniel 
> _______________________________________________
> 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 halla3 at corp.earthlink.net  Wed Sep 29 10:43:31 2004
From: halla3 at corp.earthlink.net (Andrew Hall)
Date: Wed, 29 Sep 2004 14:43:31 -0000
Subject: SUMMARY Re: jumpstart error
In-Reply-To: <1096402897.19294.1.camel@kuu.eng.atl.earthlink.net>
References: <1096402897.19294.1.camel@kuu.eng.atl.earthlink.net>
Message-ID: <1096408107.20148.26.camel@flagg>

Thanks to Darren Dunham and Eivind Nordbye for the answer.  I had to
mount up the "Operating Environment CD 1999" and run:

modify_install_server -r 2.6 -d /path/to/jumpstart/sol2.6

After a new add_install_client, and a ./check, it appears to be
installing correctly. 

Thank you,

Drew

On Tue, 2004-09-28 at 16:21, Andrew Hall wrote:
> Greetings,
> 
> I am trying to jumpstart an old Netra T1 with solaris 2.6.  I have
> everything setup but when I execute a 'boot net - install', I get the
> following error then it drops back to the ok prompt.  
> 
> krtld: error during initial load/link phase
> Fast Instruction Access MMU Miss
> 
> Any idea what is causing this and how to correct?  
> 
> Thank you for your time.
> 
> Drew
> _______________________________________________
> 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 peek at tiem.utk.edu  Wed Sep 29 12:26:47 2004
From: peek at tiem.utk.edu (Michael S. Peek)
Date: Wed, 29 Sep 2004 16:26:47 -0000
Subject: SUMMARY: [Q]: Limit on number of NFS exports? 
In-Reply-To: Your message of "Wed, 29 Sep 2004 10:36:01 EDT."
	<20040929143616.6348.qmail@mail.tiem.utk.edu> 
Message-ID: <20040929162617.8D5191E6D8@sunportal.sunmanagers.org>

> Does anyone know if there is a limit on the number of NFS exports that Solaris
> supports, and if so, is there a way to extend that number?
> 
> I have a machine (E4500 w/ Solaris 8) with lots of partitions (35 of them) but
> I can only seem to export the first 30.

Uh...  Never mind.

It helps if you make sure the entries in /etc/dfs/dfstab match the mount
points in /etc/vfstab.

Duh.

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


From RMcEwan at netopia.com  Wed Sep 29 12:29:36 2004
From: RMcEwan at netopia.com (McEwan, Ryan)
Date: Wed, 29 Sep 2004 16:29:36 -0000
Subject: SUMMARY: Weird NFS performance problems on solaris 8 and netapp
Message-ID: <F1F84A2E5F9EBD46A9BDB905EB5B103907E3AA67@mxca1.netopia.com>

Many thanks to those that replied with suggestions.  It turns out to be a
bug in the NetApp software version that I am running.
 
The Solaris 9 machine that worked great was reverse dns'ed.  The Solaris 8
machines were not.  Since they were not reverse dns'ed the netapp was
failing with a gethostbyaddr call and thus causing the slowness.  The
solaris 8 and 9 differences caused me to go down the wrong path, but
eventually truth prevailed.
 
The work around is to reverse dns the servers or add them to /etc/hosts file
on the filer itself.  They have also fixed something to prevent this in a
later release of the software.
 
Once again,
 
many thanks
 
Ryan

  _____  

From: McEwan, Ryan 
Sent: Tuesday, September 28, 2004 2:24 PM
To: 'sunmanagers at sunmanagers.org'
Subject: Weird NFS performance problems on solaris 8 and netapp


I recently installed 2 Sun V210 servers (Solaris 8 HW4 update) with 1 week
old patch recommended cluster.  I am noticing painfully slow nfs performance
to a volume mounted to a NetApp filer.  When I try to access the same netapp
volume from a Solaris 9 machine (Sun 220R with Solaris 9 U7 and 1 week old
patch cluster) I see great performance.
 
Where am I going wrong or what can I begin to check?
 
I have hard coded the /dev/bge0 settings and cisco switch to match as
follows
 
ndd -set /dev/bge0 adv_1000fdx_cap 0
ndd -set /dev/bge0 adv_1000hdx_cap 0
ndd -set /dev/bge0 adv_100fdx_cap 1
ndd -set /dev/bge0 adv_100hdx_cap 0
ndd -set /dev/bge0 adv_10fdx_cap 0
ndd -set /dev/bge0 adv_10hdx_cap 0
ndd -set /dev/bge0 adv_autoneg_cap 0
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From aielloster at gmail.com  Wed Sep 29 12:33:44 2004
From: aielloster at gmail.com (Webpro)
Date: Wed, 29 Sep 2004 16:33:44 -0000
Subject: SUMMARY: Prevent Solaris from broadcasting ARP for second
 interface over first interface.
In-Reply-To: <6e3de851040928090624bbadbd@mail.gmail.com>
References: <6e3de851040928090624bbadbd@mail.gmail.com>
Message-ID: <6e3de85104092909281b7fa792@mail.gmail.com>

I received just a couple responses.  Unfortunately, no new "holy
grail" procedures.

What I have done already is disable ARP on the interface and create
static ARP entries.  This is an intrusive procedure that must be
manually performed on each system.  Having hundreds of these pairs
makes this a challenging maintenance task.

Joe

On Tue, 28 Sep 2004 09:06:39 -0700, Webpro <aielloster at gmail.com> wrote:
> I have a server configuration where a pair or servers are connected
> via a x-over cable.  They are connected to one or more LANs via other
> interfaces.  The private x-cover connection uses addresses in the
> reserved ranges.  Servers A and B are directly connected and servers D
> and E are directly connected.  e.g.
> 
> Server A                         Server B
> Public: 210.210.210.21    Public: 210.210.210.22
>          10.10.210.21    Xover  10.10.210.22
> 
> Server D                         Server E
> Public: 210.210.210.23    Public: 210.210.210.24
>          10.10.210.21    Xover  10.10.210.22
> 
> We often see when server B x-over sends and ARP for 10.10.210.21, it
> is sent out the public interface.  Server D responds with it's private
> interface MAC address (for its 10.10.210.21 IP address).  Server A
> updates its local ARP table with the MAC from server D.  vital
> communications between Server A and B over this x-over connection are
> disrupted.
> 
> I am not running ip forwarding or routed.
> 
> I need to ensure the Solaris does not forward the ARP from the private
> link to the public interface.  I am not sure why it would since it is
> locally connected and permanent.


Joe
-- 
"Despite the hight cost of living, it still remains popular!"
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From Russell.J.LeBar at erac.com  Wed Sep 29 13:07:59 2004
From: Russell.J.LeBar at erac.com (Lebar, Russell J)
Date: Wed, 29 Sep 2004 17:07:59 -0000
Subject: SUMMARY: V480 and Qlogic Loop offline/online question
Message-ID: <0EBFC22FDE33C84B97D546599FFC4491024B7B76@STL-EXBE-I03.corp.erac.com>

Right now I am just keeping an eye on the system. I haven't seen any
more notices since that first night. Thanks for the help! -- Russ

And here are the responses I got...

Tim Chipman had this to say:
----------------------------

I've had loop-offline problems recently with our t3 (connected to a
qlogic HBA, single channel fibre, non-redundant config) which so far
have vanished since we cold-shutdown everything, unplugged all the bits,
and replugged everything, followed by clean boot and then a few hours of
SunVTS testing to be paranoid.

Some suggestions were to flash the HBA firmware ; to replace the HBA ;
and to maybe replace other components in the fibre loop if the problem
persisted.

additionally, a trawl through sunsolve found for me some hit maybe
relevant for you, todo with power-saving features spinning down internal
FCAL disks on some newer boxes such as v480 maybe (?).  Worth a look in
sunsolve if you haven't seen this yet...

And Mary Dell had this to say:
------------------------------

I have had exactly this behavior with a bad gbic - offline/online,
offline/online, sometimes for months.  I had it in an E3500 and there
was one particular kind of gbic that was very prone to this - I found a
lot of references online to a gbic from a company starting with a V I
think - vixel or something like that.  Anyway, I pulled the one gbic I
had that wasn't ibm and replaced it with an ibm and the problem went
away.  Keep in mind that usually you have to replace both gbis on that
fibre channel since there's usually not any way of telling which end of
the loop is bad - I just got "lucky" having a known-crappy part.





> -----Original Message-----
> From: sunmanagers-bounces at sunmanagers.org
> [mailto:sunmanagers-bounces at sunmanagers.org] On Behalf Of
> Lebar, Russell J
> Sent: Thursday, September 23, 2004 12:29
> To: sunmanagers at sunmanagers.org
> Subject: V480 and Qlogic Loop offline/online question
>
> We have a V480 that we started having drive problems on disk1
> (c1t1d0).
> SUN put a new backplane in the system and replaced disk1.
> Both disks are Seagate Cheetahs and of course there is the
> DVD drive. We only saw errors on disk1.
>
> I thought this took care of the problem but I noticed the
> following in the messages file:
>
> Sep 21 18:20:51 nexus6 qlc: [ID 686697 kern.info] NOTICE:
> Qlogic qlc(0):
> Loop OFFLINE
> Sep 21 18:20:52 nexus6 qlc: [ID 686697 kern.info] NOTICE:
> Qlogic qlc(0):
> Loop ONLINE
> Sep 21 18:20:56 nexus6 scsi: [ID 107833 kern.warning] WARNING:
> /pci at 9,600000/SUNW,qlc at 2/fp at 0,0/ssd at w2100000c50b8863e,0 (ssd1):
> Sep 21 18:20:56 nexus6  Unhandled Sense Key 'Vendor Unique'
> Sep 21 21:00:41 nexus6 qlc: [ID 686697 kern.info] NOTICE:
> Qlogic qlc(0):
> Loop OFFLINE
> Sep 21 21:00:45 nexus6 qlc: [ID 686697 kern.info] NOTICE:
> Qlogic qlc(0):
> Loop ONLINE
> Sep 21 21:44:26 nexus6 qlc: [ID 686697 kern.info] NOTICE:
> Qlogic qlc(0):
> Loop OFFLINE
> Sep 21 21:44:26 nexus6 qlc: [ID 686697 kern.info] NOTICE:
> Qlogic qlc(0):
> Loop ONLINE
> Sep 21 22:09:23 nexus6 qlc: [ID 686697 kern.info] NOTICE:
> Qlogic qlc(0):
> Loop OFFLINE
> Sep 21 22:09:23 nexus6 qlc: [ID 686697 kern.info] NOTICE:
> Qlogic qlc(0):
> Loop ONLINE
>
>
> By 18:20 I was restoring the filesystem (/export) from tape.
> There was about 33GB of data and the restore finished around
> 21:30. It looks like Netbackup started doing backups on
> c1t1d0s0 around 20:07  (got to love that timing).
>
> Is there still a possible problem? I have heard of GBIC
> issues with SUN systems using fibre. Running: Solaris 9 4/04
> s9s_u6wos_08a SPARC Assembled 22 March 2004.
>
> Thanks in advance!
>
>
> --
> -------------------------
> Russ LeBar
> Enterprise Rent-A-Car
> UNIX System Administrator
> _______________________________________________
> 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 kevin.m.smith at siemens.com  We