From martynas at ti.com  Mon Jan  3 03:32:52 2005
From: martynas at ti.com (Buozis, Martynas)
Date: Mon, 3 Jan 2005 09:32:52 +0100
Subject: SUMMARY: Solaris problems
Message-ID: <8360475C2725774B8E7D37D5B04356A102177EBE@dfre2k03.itg.ti.com>

Hello

With SUN support help this problem was solved with Solaris patch
114600-04 that fixed following bug :

Bug ID: 4730657
Synopsis: Maximum MTU of VLAN interface is limited to 1496 bytes.
Category: ethernet_vlan
Subcategory: software
State: 11-Closed
Description:
The vlan kernel module reduces the value of dl_max_sdu by the size of a
VLAN tag.
The result of this is that a vlan interface will have an maximum MTU of
1496 bytes.
The maximum MTU value should be 1500 bytes.
Work Around:
None.
Fixed by patch:
Integrated in Build: 1.0prefcs
Duplicate of:
Related Change Request(s):
4823410
Date Modified: 03/27/2004 02:45:00 GMT+00:00
Public Summary:
The vlan kernel module reduces the value of dl_max_sdu by the size of a
VLAN tag.
The result of this is that a vlan interface will have an maximum MTU of
1496 bytes.
The maximum MTU value should be 1500 bytes.
e.g.
anon(5.9)$ ifconfig -a
lo0: flags=1000849<UP,LOOPBACK,RUNNING,MULTICAST,IPv4> mtu 8232 index 1
        inet 127.0.0.1 netmask ff000000
ce101000: flags=1000842<BROADCAST,RUNNING,MULTICAST,IPv4> mtu 1496 index
3
        inet 0.0.0.0 netmask 0
anon(5.9)$


With best regards
Martynas Buozis


-----Original Message-----
From: sunmanagers-bounces at sunmanagers.org
[mailto:sunmanagers-bounces at sunmanagers.org] On Behalf Of Buozis,
Martynas
Sent: Friday, December 31, 2004 4:45 AM
To: sunmanagers at sunmanagers.org
Subject: FW: Solaris problems

Hello

Please help me. I have serious problems and not sure were is actual
problem. I have NAT'ed connection that is handled by SUN GigaSwift card
with tagged VLAN  - NIC ce700000. When I open connection to host
everything works well for a while and after some packets ICMP errors
starts to come out :

nickel.internal.ffab.tide.ti.com -> ins6         XWIN C port=43680
        ins6 -> nickel.internal.ffab.tide.ti.com ICMP Destination
unreachable (Needed to fragment: next hop MTU = 1496)
nickel.internal.ffab.tide.ti.com -> ins6         XWIN C port=43680
        ins6 -> nickel.internal.ffab.tide.ti.com ICMP Destination
unreachable (Needed to fragment: next hop MTU = 1496)
nickel.internal.ffab.tide.ti.com -> ins6         XWIN C port=43680
        ins6 -> nickel.internal.ffab.tide.ti.com ICMP Destination
unreachable (Needed to fragment: next hop MTU = 1496)

1496 MTU is for ce70000 card. Is this problem IPFilter related or it is
Solaris problem ?

Please help me and respond ASAP. Thank you for your help.

With best regards
Martynas
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From alex at indo.com  Mon Jan  3 23:16:37 2005
From: alex at indo.com (Alexander Surbakti)
Date: Tue, 04 Jan 2005 11:16:37 +0700
Subject: How to send email using perl
Message-ID: <41DA18A5.3020104@indo.com>

I try to search in google and I found this script with some of modify:

#!/usr/bin/perl
use CGI;

$send_to = param('send_to');

if ($send_to eq '') {
   
<FORM method="POST" action="?">
<INPUT type="text" name="send_to">
<INPUT type="submit">
</FORM>
}
else {
my $query    = new CGI;
my $sendmail = "/usr/sbin/sendmail -t";
my $reply_to = "Reply-to: alex at indo.com";
my $subject  = "Subject: Confirmation of your submission";
my $content  = "Thanks for your submission.";
my $to       = $query->param('send_to');
my $file     = "subscribers.txt";

unless ($to) {
  print $query->header;
  print "Please fill in your email and try again";
}

open (FILE, ">>$file") or die "Cannot open $file: $!";
print $to,"\n";
close(FILE); 

my $send_to  = "To: ".$query->param('send_to');

open(SENDMAIL, "|$sendmail") or die "Cannot open $sendmail: $!";
print SENDMAIL $reply_to;
print SENDMAIL $subject;
print SENDMAIL $to;
print SENDMAIL "Content-type: text/plain\n\n";
print SENDMAIL $content;
close(SENDMAIL);

print $query->header;
print "Confirmation of your submission will be emailed to you.";
}


The script I named with sendingmail.pl. When I run the script it just 
print  $to which is the email address I typed in form before submited 
and no email come in $to inbox.
As an addition, my perl script put in different server with our mail 
server. sendingmail.pl I put in cgi-bin at training.indo.com while our 
mail server is mail.indo.com. My question is what should I do with 
$sendmail? It seems that the path is belongs to sendmail in mail server.

thanks,
-alex-


From fbertels at ixis-cib.com  Tue Jan  4 05:43:09 2005
From: fbertels at ixis-cib.com (Bertels, Frank)
Date: Tue, 4 Jan 2005 11:43:09 +0100
Subject: [Summary] Patchpro on sol9 cannot download patches - Error,
 caught No valid download source
Message-ID: <7CDEE32685A846489AD1E72D1C35EEEE020317BE@msbac.cm.net>

The solution was:  remove of an older j2sdk and installation of actual
j2sdk-1_4_2_06 Package (and installation of the j2sdk solaris patches,
too).

(But I don't know if in the meanwhile SUN has changed anything on their
website.)

________________________________

From: Bertels, Frank
Sent: Friday, December 10, 2004 4:49 PM
To: 'sunmanagers at sunmanagers.org'
Subject: Patchpro on sol9 cannot download patches - Error, caught No
valid download source


Hi,

I want to use a Ultra1 (Solaris 9, installed with newest Recommended
Patch Cluster) with Patch Pro 2.2.
I installed pproSunOSsparc5.9jre2.2.tar.gz and the appropriate patch
116126-03, and I checked my proxy configuration.
Everything should be OK, but if I execute patchpro I get the following
error:

root at toronto:/(517) > /opt/SUNWppro/bin/pprosvc -l
Assessing patches needed for your system...
...Cannot establish a network connection with the patch server
("https://patchpro.sun.com/database/pprodetectors.jar"). Check your
proxy configuration or contact your local system administrator for
assistance.
Cannot establish a network connection with the patch server
("https://patchpro.sun.com/database/patchprodb.zip"). Check your proxy
configuration or contact your local system administrator for assistance.
.Problem detected. Refer to the log (Location: System log file) for
detailed information about the problem.
.
Retrieved patch list...
No patches required.

If I look to /var/adm/messages, I find:

Dec 10 15:53:31 toronto root: [ID 702911 user.crit] Fri Dec 10 15:53:30
MET 2004(CRITICAL) =>
com.sun.patchpro.model.PatchProStateMachine at 17725c4
<mailto:com.sun.patchpro.model.PatchProStateMachine at 17725c4>
<=downloadDetectors.run(): Error, caught No valid download source..
Dec 10 15:53:31 toronto root: [ID 702911 user.crit] Fri Dec 10 15:53:31
MET 2004(CRITICAL) =>
com.sun.patchpro.database.GroupPatchDBBuilder at 1506dc4
<mailto:com.sun.patchpro.database.GroupPatchDBBuilder at 1506dc4>
<=GroupPatchDBBuilder: Internal error. Build failed.
Dec 10 15:53:31 toronto root: [ID 702911 user.crit] Fri Dec 10 15:53:31
MET 2004(CRITICAL) =>
com.sun.patchpro.database.GroupPatchDBBuilder at 1506dc4
<mailto:com.sun.patchpro.database.GroupPatchDBBuilder at 1506dc4>
<=com.sun.patchpro.database.DBBuilderFailedException: No valid download
source.
Dec 10 15:53:31 toronto         at
com.sun.patchpro.database.MemoryPatchDBBuilder.buildDB(MemoryPatchDBBuil
der.java:118)
Dec 10 15:53:31 toronto         at
com.sun.patchpro.database.MemoryPatchDBBuilder.
Dec 10 15:53:31 toronto root: [ID 702911 user.crit] Fri Dec 10 15:53:31
MET 2004(CRITICAL) => com.sun.patchpro.model.PatchProStateMachine at a761fe
<mailto:com.sun.patchpro.model.PatchProStateMachine at a761fe>
<=downloadPatchDB.run(): Error, caught No valid download source..
Dec 10 15:53:32 toronto root: [ID 702911 user.crit] Fri Dec 10 15:53:31
MET 2004(CRITICAL) =>
com.sun.patchpro.model.PatchProStateMachine at 17725c4
<mailto:com.sun.patchpro.model.PatchProStateMachine at 17725c4>
<=java.io.IOException: No valid download source.
Dec 10 15:53:32 toronto         at
com.sun.patchpro.util.CachingDownloader.setSourceURL(CachingDow
Dec 10 15:53:32 toronto root: [ID 702911 user.crit] Fri Dec 10 15:53:32
MET 2004(CRITICAL) => com.sun.patchpro.model.PatchProStateMachine at a761fe
<mailto:com.sun.patchpro.model.PatchProStateMachine at a761fe>
<=com.sun.patchpro.database.DBBuilderFailedException: No valid download
source.
Dec 10 15:53:32 toronto         at
com.sun.patchpro.database.GroupPatchDBBuilder.buildDB(GroupPadmesgtchDBB
uilder.java:122)
Dec 10 15:53:32 toronto         at
com.sun.patchpro.model.PatchProModel.downloadPatchDB(PatchProModel.java:
1967)
Dec 10 15:53:32 toronto         at
com.sun.patchpro.model.PatchProStateMachine.run(PatchProStateMachine.jav
a:339)
Dec 10 15:53:32 toronto         at
com.sun.patchpro.util.State.run(State.java:275)
Dec 10 15:53:32 toronto         at java.lang.Thread.

Does anybody have an idea to solce the problem?
(I installed the same on a Solaris8 machine and there it's running.)
Here's my conf:
root at toronto:/(518) > /opt/SUNWppro/bin/pprosetup -L
Log file location:      System log file (consult /etc/syslog.conf for
exact location)
Log file size:          50000
Download directory:     /var/sadm/spool
Sequester directory:    /var/sadm/spool/patchproSequester
Proxy server name:      proxyserver
Proxy server port:      8080
Server URL:
https://patchpro.sun.com/servlet/com.sun.patchpro.server.PatchProServerS
ervlet/
Database URL:           https://patchpro.sun.com/database/patchprodb.zip
Detectors URL:
https://patchpro.sun.com/database/pprodetectors.jar




________________________________________________________________________

Frank Bertels

Email: fbertels at ixis-cib.com <mailto:fbertels at ixis-cib.com>

________________________________________________________________________

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

This message and any attachments are confidential and intended solely for the
addressees, it may contain confidential and/or privileged information. Any use
not in accord with its purpose, any dissemination or disclosure, either whole
or partial, is prohibited except formal approval. If you receive this message
in error, please delete it and immediately notify the sender. Neither IXIS CIB
nor IXIS Corporate & Investment Bank shall be liable for the message if
altered, changed or falsified.
--------------------------------------------------------
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From srv_hassan at yahoo.co.in  Wed Jan  5 09:25:06 2005
From: srv_hassan at yahoo.co.in (Viveka Shankara)
Date: Wed, 5 Jan 2005 14:25:06 +0000 (GMT)
Subject: How to import messages from netscape 4.78 to Mozilla 1.4
Message-ID: <20050105142506.53945.qmail@web8309.mail.in.yahoo.com>

Hello,

How to import Netscape 4.78 mails to Mozilla 1.4.
I have solaris 9 and tried by selecting the import
option in the mozilla tools, but it didn't worked out,
Please guide me how to process, as this is very
critical for my work.

Thanks,


________________________________________________________________________
Yahoo! India Matrimony: Find your life partner online
Go to: http://yahoo.shaadi.com/india-matrimony


From bshah at citadon.com  Tue Jan  4 13:14:01 2005
From: bshah at citadon.com (Bhavesh Shah)
Date: Tue, 4 Jan 2005 10:14:01 -0800
Subject: SUMMARY: Cluster for Solaris 
Message-ID: <0FEA43415756E44AB368AC58C1A4F4F301F9E075@a4mail2k01.citadoncorp.com>

Hi,
I am really thankful to all for their prompt responses.
Many thanks to:
Alfredo De Luca
Hans Jacobsen
Reggie Beavers
McEwan Ryan
Angelo McComis
Alan Pae
Neil Quiogue
Ragesh Kumar

I got mixed responses. Most of the suggestions were to use V/O/S
(Veritas/Oracle/Sun) combination is really first-rate
For Oracle DB most suggestions were to use oracle 10g cluster which is
in-built.
As regards Sun and Veritas Clusters Sun is very difficult to setup but
once its setup its robust and reliable but requires Sun professional to
install and configure where as Veritas is easy to setup and administer.
Most responses were to use Veritas Cluster server. The only problem with
Veritas is its cost.
For FREEWARE there are tow suggestions:
HA/FST at http://www.fstha.com which also has free support.
Other is http://linux-ha.org

I have still not decided which way to go but definitely try freeware
before deciding over commercial ones.

Thanks a lot guys
B

-----Original Message-----
 Hi Gurus,

We are planning buy Cluster software for Oracle DB as well as File
system on Sun Solaris 8/9. So far I came across two well known in
markets:

VCS and Sun Cluster.

Can you please share your opinion/experience which will be better in
terms of Cost, reliable and efficient.

 Also if you are using some FREEWARE which is equally good and tested, I
will prefer that.

TIA

B
_______________________________________________
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 hareshgc at rediffmail.com  Wed Jan  5 00:26:49 2005
From: hareshgc at rediffmail.com (Haresh G. Chandnani)
Date: 5 Jan 2005 05:26:49 -0000
Subject: SUMMARY: File Creation Time UFS
Message-ID: <20050105052649.14609.qmail@webmail26.rediffmail.com>

Thank you to all those who replied. From all the replies & search I found
there is no way to find out when the file was created in UFS.

Regards
Haresh G. Chandnani
http://haresh.chandnani.tripod.com/index.html
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From solaris at be.itu.edu.tr  Wed Jan  5 09:04:31 2005
From: solaris at be.itu.edu.tr (Ayhan Kanmaz)
Date: Wed, 5 Jan 2005 16:04:31 +0200 (EET)
Subject: SUMMARY: User Level Accounting
In-Reply-To: <Pine.LNX.4.58.0501041534390.12329@eyup.be.itu.edu.tr>
References: <Pine.LNX.4.58.0501041534390.12329@eyup.be.itu.edu.tr>
Message-ID: <Pine.LNX.4.58.0501051556200.31581@silivri.be.itu.edu.tr>

Hello,

I thanked all who replied. The suggestions are

- orca  http://www.orcaware.com/orca/ - open source project
- ovoux http://www.managementsoftware.hp.com/products/ovoux/ - product

I will try to use orca.

my original post  was 

On Tue, 4 Jan 2005, Ayhan Kanmaz wrote:

> Hello,
> 
> I have a problem with accounting. I managed Sunfire 12k and 
> 24 v210 cluster system. My manager wants me some user level 
> reports showing cputime,memory, etc... dayly,monthly,
> yearly, etc... . I enabled process accounting. However, i do not know
> how i create reports and visualize them. I googled before, but all
> documents I found were written how i can activate the accounting.
> 
> Can you indicate me any valuable links, documents,tools or ideas about 
> generating accounting reports and visualizing them?   
> 
> OS Version: Solaris 9
> _______________________________________________

With regards,
Ayhan
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From ghenry at cmi.univ-mrs.fr  Wed Jan  5 09:33:36 2005
From: ghenry at cmi.univ-mrs.fr (Gerard Henry)
Date: Wed, 05 Jan 2005 15:33:36 +0100
Subject: SUMMARY: unable to quit gv!
In-Reply-To: <41D95BC6.9050702@cmi.univ-mrs.fr>
References: <41D95BC6.9050702@cmi.univ-mrs.fr>
Message-ID: <41DBFAC0.5090509@cmi.univ-mrs.fr>

thanks to rene occelli

the issue was in PATH
if /usr/bin is set before /usr/local/bin, gv freeze when trying quit
don't understand why...
but did a script to set correctly PATH
PATH=/usr/local/bin:/usr/bin




>
>
> unable to quit gv (postscript and pdf previewer) since one week
> i launch gv with a postscript file
> gv toto.ps
> changing pages, but when i try to quit, windows freeze, i have to do 
> CTRL-C to quit (window seems to be not repainted)
>
> if i open gv without a file, i can quit
> if i use only gs, there is no problem
>
> i don't understand where is the problem, if anybody has a suggestion?
>
> i'm using gv from sunfreeware packages
> gv: 3.5.8
> gs: 8.00
> xaw3d: 1.5
> solaris 9 just patched with last recommended cluster of patches
>
> i compare shared used libraries with another machine where the same 
> gv's release works well
>
> thanks in advance,
>
> gerard
> _______________________________________________
> 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 carolyn at cs.usna.edu  Wed Jan  5 07:46:48 2005
From: carolyn at cs.usna.edu (Carolyn Mayr {ITSD/CS})
Date: Wed, 5 Jan 2005 07:46:48 -0500 (EST)
Subject: SUMMARY: RE: Can't open boot device (need to fsck)
Message-ID: <200501051248.j05CmL7b026715@rattler.cs.usna.edu>

Thank you to everyone who came to my rescue this morning.
    Simon Loewenthal
    David Harrington
    Allen Wooden
    Rene Occelli
    
Lots of good information to narrow down the problem.
Eugene Schmidt had the final answer that worked.  The
SunBlade 1500s only work with the Solaris 9 4/04 CD.

Yesterday I used all my Solaris 9 CD (2002-2004) on my
SunBlade 150s and they all worked fine.  That's where
the confusion happened.  Thank you all!  
				Carolyn Mayr
				
------------- Begin Forwarded Message -------------
From: "Eugene Schmidt" <fereug at acute.co.za>
To: "Carolyn Mayr {ITSD/CS}" <carolyn at cs.usna.edu>
Subject: RE: Can't open boot device (need to fsck)
Date: Wed, 5 Jan 2005 14:31:50 +0200


Hi

The boot device referred to here is  the CD-Rom disk.

I sugegst use CD 1 of 2 from the install set. ( I suspect you may be using
the wrong one ?)

Also ensure you are at least using SOlaris 9 4/04 or later.

Best of luck

regards

E

-----Original Message-----
From: sunmanagers-bounces at sunmanagers.org
[mailto:sunmanagers-bounces at sunmanagers.org]On Behalf Of Carolyn Mayr
{ITSD/CS}
Sent: 05 January 2005 02:14
To: sunmanagers at sunmanagers.org
Subject: Can't open boot device (need to fsck)


Help!

I trashed my /etc/passwd file on my SunBlade 1500s and am trying
to boot from my Solaris 9 installation CD.  The system was frozen so
I had to do a Stop-A then

>boot cdrom -s

I'm getting this error:

   Rebooting with command: boot cdrom -s
   Boot device: /pci at 13,600000/ide at d/cdrom at 2,0:f  File and args: -s
   Cant' read disk label.
   Can't open disk label package

   Can't open boot device

The problem is I need to do an fsck on the / system which got
corrupted.  That's probably why it can't read the boot device.
Remember I don't have a root password so I can't type the
password to fsck the Sun.  I tried powering on/off and >boot -r
but still the same error.  Help!


=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Carolyn Mayr (IT Specialist/UNIX sysadm) Voice: 410-293-6808 (sec-6800)
Computer Science Department, DivMath&Sci Email: carolyn at cs.usna.edu
572M Holloway Road, Michelson Hall, 9F   FAX:   410-293-2686
U.S. Naval Academy                       WWW:   http://www.cs.usna.edu
Annapolis, MD  21402-5002                USNA:  410-293-1000
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


------------- End Forwarded Message -------------


=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Carolyn Mayr (IT Specialist/UNIX sysadm) Voice: 410-293-6808 (sec-6800)
Computer Science Department, DivMath&Sci Email: carolyn at cs.usna.edu     
572M Holloway Road, Michelson Hall, 9F   FAX:   410-293-2686
U.S. Naval Academy                       WWW:   http://www.cs.usna.edu
Annapolis, MD  21402-5002                USNA:  410-293-1000
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From glowe at sbcglobal.net  Wed Jan  5 12:41:24 2005
From: glowe at sbcglobal.net (Grant Lowe)
Date: Wed, 5 Jan 2005 09:41:24 -0800 (PST)
Subject: SUMMARY (late): Jumpstart problems
Message-ID: <20050105174124.92332.qmail@web81703.mail.yahoo.com>

Sorry for the late reply, but have had lots of stuff going on with the holidays.  Anyway, I got the problem resolved over at BigAdmin at Sun.  There were many various answer, which partly solved the problem.  But the user hassmaschine over there helped me solve the problem by putting all the pieces together for me.  The answer was this:the netmask was the problem. The netmask I was using. The 255.255.248.0 netmask was causing broadcast problems. Making the mask much smaller, to a 255.255.255.248, made it work, and also made the IP traffic move much faster as well.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From duprec at scorec.rpi.edu  Wed Jan  5 16:13:58 2005
From: duprec at scorec.rpi.edu (Christophe Dupre)
Date: Wed, 5 Jan 2005 16:13:58 -0500 (EST)
Subject: SUMMARY: Large disk in Netra X1 ?
In-Reply-To: <Pine.GSO.4.58.0501051536140.11836@sinclair.scorec.rpi.edu>
References: <Pine.GSO.4.58.0501051536140.11836@sinclair.scorec.rpi.edu>
Message-ID: <Pine.GSO.4.58.0501051609590.11836@sinclair.scorec.rpi.edu>

Thanks to Tim Chapman for the pointer that allowed me to google for the
complete answer. Thanks to:
Edmunds Chris
Mao YS
Roland Merk
Juergen Waiblinger

for letting me know they were out of office.

This thread on the Sun support forums has the complete information on how
to use format to make the X1 recognize 137GB:
http://supportforum.sun.com/hardware/index.php?t=msg&goto=7912&rid=0#msg_7912

The rest of the disk cannot be used as the IDE controller doesn't support
more than 24-bit addressing. This is good enough for my needs.

On Wed, 5 Jan 2005, Christophe Dupre wrote:

> I'm trying to upgrade the internal disks of a Netra X1 to use as an FTP
> server. I took out the factory 20GB disk and replaced it with a new 160GB
> disk (Seagate ST3160023A). The disk is seen properly from openboot:
> ok probe-ide
>   Device 0  ( Primary Master )
>           ATA Model: ST3160023A
>
> but format sees the disk as having only 0.5GB:
> Primary label contents:
>
> Volume name = <        >
> ascii name  = <ST3160023A cyl 255 alt 2 hd 16 sec 255>
> pcyl        =  257
> ncyl        =  255
> acyl        =    2
> nhead       =   16
> nsect       =  255
> Part      Tag    Flag     Cylinders       Size            Blocks
>   0 unassigned    wm       0              0         (0/0/0)         0
>   1 unassigned    wm       0              0         (0/0/0)         0
>   2     backup    wu       0 - 254      508.01MB    (255/0/0) 1040400
>   3 unassigned    wm       0              0         (0/0/0)         0
>   4 unassigned    wm       0              0         (0/0/0)         0
>   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
>
> I have several V100 with disks upgraded to 80GB without problem, so what's
> going on with this one ?
>
>
>
> --
> Christophe Dupre
> System Administrator, Scientific Computation Research Center
> Rensselaer Polytechnic Institute
> Troy, NY        USA
> Phone: (518) 276-2578  -  Fax: (518) 276-4886
> _______________________________________________
> sunmanagers mailing list
> sunmanagers at sunmanagers.org
> http://www.sunmanagers.org/mailman/listinfo/sunmanagers
>
>


--
Christophe Dupre
System Administrator, Scientific Computation Research Center
Rensselaer Polytechnic Institute
Troy, NY        USA
Phone: (518) 276-2578  -  Fax: (518) 276-4886
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From lupe at lupe-christoph.de  Thu Jan  6 04:50:15 2005
From: lupe at lupe-christoph.de (Lupe Christoph)
Date: Thu, 6 Jan 2005 10:50:15 +0100
Subject: SUMMARY: Experiences with multi-Terabyte filesystems
In-Reply-To: <20050104090534.GA25444@lupe-christoph.de>
References: <20050104090534.GA25444@lupe-christoph.de>
Message-ID: <20050106095015.GA15677@lupe-christoph.de>

On Tuesday, 2005-01-04 at 10:05:34 +0100, Lupe Christoph wrote:

> ...

> So I would like to ask for your experiences with filesystems larger than
> 4 TB.

Thanks to:
Bousquet Francois
Jason Shatzkamer
Mao YS-qch1521
Juergen Waiblinger
Sirisena, Navi
Roland Merk
Ballesteros, Dave D
Stott, Trevor  xt55302

for letting me know they were out of office. I have forwarded their
mailbox addresses to well-known Spammers. Thanks for the money, folks!
(Just kidding.)

Petri Kallberg and Drew Skinner pointed me to the upcoming ZFS
(Zettabyte File System). But ZFS will not be in the initial release of
Solaris 10, but MU2 or so. And I would like to see somebody else test
this thoroughly before I commit to it. My client will not pay me for
extensive tests, alas.

Adam Levin said "we implemented some Network Appliance storage systems
with multi-terabyte filesystems." Netapp filers are even more expensive
than the SamFS license my client would have to purchase.

Drew Skinner mentioned "I have 24TB (2 x 12TB) filesystems with SAMfs
and have absolutely no problems with it." Good to know, I'm similarly
satisfied with SamFS, but the client wants something cheap. :-(

Doug Hughes said "not a problem with something like VxFS. Less of a
problem with UFS+ with logging turned on, but VxFS has a marginal edge
with larger sizes.  The larger you get, the faster it gets for crash
recovery (in comparison).  It goes up to 32TB and many sites use the
whole thing (or more, depending upon OS version and Veritas version)"

Finally, Filipe Litaiff asked me to summarize, which I have just done.

So I have to conclude that such large filesystems are indeed a challenge
few people have yet faced. And even fewer would trust UFS with that.

Thank you all.
Lupe Christoph
-- 
| lupe at lupe-christoph.de       |           http://www.lupe-christoph.de/ |
| Ask not what your computer can do for you                              |
| ask what you can do for your computer.                                 |
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From Y.Ng at agi.rws.minvenw.nl  Thu Jan  6 04:52:25 2005
From: Y.Ng at agi.rws.minvenw.nl (Ng, Y. (Yatsen))
Date: Thu, 6 Jan 2005 10:52:25 +0100
Subject: SUMMARY: show scsi devices
Message-ID: <1FEE21DC1054D5119FCC0002A5517B2403E119FF@rws-mdi-dt2-01.agi.rws.minvenw.nl>

Thanks to Anthony D'Atri.  

iostat -E

Disclaimer
************************************************************************
Aan dit bericht kunnen geen rechten worden ontleend. Dit bericht is
uitsluitend bestemd voor de geadresseerde. Als u dit bericht per abuis
hebt ontvangen, wordt u verzocht het te vernietigen en de afzender te 
informeren. Wij adviseren u om bij twijfel over de juistheid of de 
volledigheid van de mail contact met afzender op te nemen.

This message shall not constitute any rights or obligations.
This message is intended solely for the addressee.
If you have received this message in error, please delete it and
notify the sender immediately. When in doubt whether this message
is correct or complete, please contact the sender.
************************************************************************
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From JWalton at alldata.net  Thu Jan  6 11:14:41 2005
From: JWalton at alldata.net (Walton, Joe)
Date: Thu, 6 Jan 2005 11:14:41 -0500
Subject: SUMMARY: Solaris 9 Certification Practice Exams
Message-ID: <A0A3E2CF6CF60C46A5E8B3ED8E014A9816E4A3@scad1exch02.corp.alldata.net>

Thank you all who responded.

Here is what I got.

These sites have practice exams for sale:
www.unixed.com
www.cramsession.com
www.sun.com

This site has a free exam which is mostly a general Solaris exam:
www.sunguru.com <http://www.sunguru.com/>

I also was notified that ebay sells practice exams for around $10.  This
is true.




Joe Walton
Unix System Administrator
Office Phone: 614-944-5863
Cell Phone: 614-657-7608
Email:  jwalton at alldata.net <mailto:jwalton at alldata.net>
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From martini at raider.llnl.gov  Fri Jan  7 13:34:33 2005
From: martini at raider.llnl.gov (Dave Martini 1)
Date: Fri, 7 Jan 2005 10:34:33 -0800 (PST)
Subject: SUMMARY: Resizing a Veritas Volume
Message-ID: <200501071834.j07IYWXE023228@raider.llnl.gov>

Thanks to all the replies.

I tried growfs but that did not work for me.

vxresize apparently is the right command

/usr/lib/vxvm/bin/vxresize data1 30g

But I got an error when trying this

ERROR: Problem running mkfs command for volume data1, in diskgroup del8


So I'm going to backup the data and recreate the volume manually
and then restore the data.

Dave Martini
LLNL

My original question:

I am running Veritas Volume Manager 3.0.4. I went into the GUI
and selected a volume and under actions I selected resize volume 
and increased it by 5 gigs. However, the file system itself did not 
grow. Anyone know the command to grow the file system as well?
growfs, tunefs, mkfs????
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From sashah76 at gmail.com  Sun Jan  9 21:14:48 2005
From: sashah76 at gmail.com (solaris)
Date: Sun, 9 Jan 2005 21:14:48 -0500
Subject: SUMMARY: Reinstall solaris on sun ultra 5
Message-ID: <f6727b9a05010918146f7e5c5b@mail.gmail.com>

Hello all,

In reply to my problem:

Michael Horton suggested connecting the machine to a small hub/switch
and then trying to boot the machine of the solaris sw cd.

It worked, I was then able to install solaris from scratch thus
avoiding any previous passwords that were set.

However, I found no work around for the obp password, I think only sun
can get around that.

Many thanks to all.

Sashah.

*********
There are only 10 types of people who understand binary, those who do
& those who dont.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From ahoesch at smartsoft.de  Mon Jan 10 06:11:08 2005
From: ahoesch at smartsoft.de (=?ISO-8859-1?Q?Andreas_H=F6schler?=)
Date: Mon, 10 Jan 2005 12:11:08 +0100
Subject: Summary: strftime on Solaris
In-Reply-To: <5D5DD938-602A-11D9-BE23-000393CA0072@smartsoft.de>
Message-ID: <541973EA-62F8-11D9-9108-000393CA0072@smartsoft.de>

Dear managers,

thanks to

crist.clark at globalstar.com
JV <jv711 at yahoo.com>
Hutin Bertrand <Bertrand.Hutin at fr.Fujitsu.com>
Lars Hecking <lhecking at users.sourceforge.net>

and especially to

Koef <koef at notsupported.org>

Solaris includes a strftime() function. It seems that the test for 
strftime in ./configure of hylafax is bogus.

int t() { strftime(); return 0; }
int main(){ t(); return 0; }

Koef suggested to comment out the test for strftime what I did. I had a 
similar problem with the TIFF library,so I comemnted this test out as 
well and then was able to build hylafax on Solaris 9 with ggcc 3.4.x. 
It was reported that these problems do not occur with gcc.3.x. However, 
I haven't verified that.

Thanks a lot!

Regards,

   Andreas

Original question:
> I am currently trying to build hylafax 4.2.0 on Solaris 9. Configuring 
> this package gives me
>
> No strftime function located in system libraries.
>
> The strftime function used to format date&time information was not 
> located
> in the normal C library or in the -lintl library (present on some 
> systems).
> This software expects this to be available and will not compile 
> without it.
> If this function is located in a non-standard location you can specify 
> the
> library in the MACHDEPLIBS configuration parameter.  Otherwise you 
> should
> figure out why the usual testing done by this script failed; consult 
> the
> diagnostics in the config.log file for more information.
>
> Unrecoverable error!  Once you've corrected the problem rerun this 
> script.
>
> Solaris contains a strftime function. Any idea how to convince the 
> configure script to find(use) it?
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From nskyrca at syr.edu  Mon Jan 10 10:19:15 2005
From: nskyrca at syr.edu (Nicole Skyrca)
Date: Mon, 10 Jan 2005 10:19:15 -0500
Subject: SUMMARY: SAN question and cfgadm
Message-ID: <s1e256b8.001@gwia201.syr.edu>

Sorry for the late summary.

I ended up using IBM's Subsystem Device Driver (IBMsdd) software
instead of Sun's mpxio for the multipathing support.  This software
made it easy for me to identify which luns I wanted removed from
the SAN using addresses that our storage administrators could
see on the SAN.

To configure the V440 to see the new disks, I ran
"cfgadm -c configure c#" for each controller. Then I had
to run some IBMsdd  commands to get the IBMsdd drivers to recognize
the disks.  Then I created disksuite devices using the IBMsdd names
for the luns.

In order to remove the luns used for the temporary filesystem from
the V440, I unmounted the filesystem, ran metaclear on the
metadevice name.  Then, I used IBMsdd to determine the addresses
of the luns I wanted removed, and asked the storage administrators
to remove the luns.  They must remove the luns before you unconfigure
them from the  machine.

After the luns are removed,run the following command.  The luns that were
removed show up as "unusable".

cfgadm *o show_FCP_dev *al

Example output:
Ap_Id                          Type         Receptacle   Occupant
Condition
c2::5005076300c39cea,0         unavailable  connected    configured
unusable
c2::5005076300c39cea,1         unavailable  connected    configured
unusable
c2::5005076300c39cea,2         unavailable  connected    configured
unusable
c2::5005076300c39cea,3         disk         connected    configured   unknown
c2::5005076300c39cea,4         disk         connected    configured   unknown
c2::5005076300c39cea,5         disk         connected    configured   unknown
c2::5005076300c39cea,6         disk         connected    configured   unknown


Next, unconfigure the LUNS that were removed from the SAN by running the
following
command for each WWN that has a Condition status of "unusable".

cfgadm *o unusable_FCP_dev *c unconfigure  cx::WWN
where "x" is the controller number and WWN is the world wide number

Example :
cfgadm -o unusable_FCP_dev -c unconfigure c2::5005076300c39cea

After removing the luns, I had to run a few more IBMsdd commands so IBMsdd
will no longer see the devices.


One last note, in order to see the luns that were removed as "unusable" with
the cfgadm command, you have to install the Sun San Foundation Kit using
the install_it script.  I had originally installed only portions of the SF
kit, and
it wasn't working properly.

Thank you to everyone who replied to this post!!
Nicole




>>> "Nicole Skyrca" <nskyrca at syr.edu> 10/27/2004 2:15:06 PM >>>
Hello,
I have a Sunfire v440 connected to an IBM ESS (shark) model 2150 with
McData Intrepid 6040 Director switch via two fiber channel connections.

I'm using Suns Traffic Manager Software (mpxio) for the multipathing
support.  I configured
a 150 GB disk partition using a DiskSuite concatination.

After I configured this 150 GB disk partition, another 250 GB of disk
from the SAN has been assigned
to this v440 that will be used for different filesystem on a temporary
basis.

How do I get the v440 to see this new disk space without destroying the
original 150 GB
partition that is already setup?  How do I unconfigure the extra 250 GB
of disk without
destroying the original 150 GB partition? Will cfgadm destroy the
original partition?

Thank you!
Nicole Skyrca

PS Does anyone have any good documentation about SANs configuration and
troubleshooting, or
on cfgadm?
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From ahoesch at smartsoft.de  Mon Jan 10 10:45:26 2005
From: ahoesch at smartsoft.de (=?ISO-8859-1?Q?Andreas_H=F6schler?=)
Date: Mon, 10 Jan 2005 16:45:26 +0100
Subject: Summary: Apache 2.x on Solaris 9
In-Reply-To: <78FF0A15-631D-11D9-9108-000393CA0072@smartsoft.de>
Message-ID: <A5F15528-631E-11D9-9108-000393CA0072@smartsoft.de>

Dear managers,

thanks to

"Imrick, Michael (GE Consumer & Industrial)" <Michael.Imrick at ge.com>
"Poulter, Dale" <dale.poulter at vanderbilt.edu>

for their prompt response. Both asked me to check whether the GROUP was 
correctly set in httpd.conf. I set it to

Group nobody

and the beast started working.

Regards,

   Andraes

Original question:
> I am trying to get apache-2.0.52 from sunfreeware.com to work on a 
> Solaris 9 box. When I do
>
> 	/usr/local/apache2/bin/apachectl start
>
> I get the following in /usr/local/apache2/logs/error_log.
>
> [Mon Jan 10 16:31:18 2005] [warn] pid file 
> /usr/local/apache2/logs/httpd.pid overwritten -- Unclean shutdown of 
> previous Apache run?
> [Mon Jan 10 16:31:18 2005] [alert] (22)Invalid argument: setgid: 
> unable to set group id to Group 4294967295
> [Mon Jan 10 16:31:18 2005] [alert] (22)Invalid argument: setgid: 
> unable to set group id to Group 4294967295
> [Mon Jan 10 16:31:18 2005] [alert] (22)Invalid argument: setgid: 
> unable to set group id to Group 4294967295
> [Mon Jan 10 16:31:18 2005] [alert] (22)Invalid argument: setgid: 
> unable to set group id to Group 4294967295
> [Mon Jan 10 16:31:18 2005] [alert] (22)Invalid argument: setgid: 
> unable to set group id to Group 4294967295
> [Mon Jan 10 16:31:18 2005] [notice] Apache/2.0.52 (Unix) DAV/2 
> SVN/1.0.4 configured -- resuming normal operations
> [Mon Jan 10 16:31:18 2005] [alert] Child 16023 returned a Fatal 
> error...\nApache is exiting!
>
> I am puzzled. What does this message try to tell me?
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From sudhakar.peram at corpusinc.com  Mon Jan 10 14:14:37 2005
From: sudhakar.peram at corpusinc.com (Sudhakar)
Date: Mon, 10 Jan 2005 13:14:37 -0600
Subject: SUMMARY :Issue with cron for a particular user
References: <20050110173432.70575.qmail@web54210.mail.yahoo.com>
Message-ID: <042f01c4f748$a0c1fbf0$be9c8071@its.verizon.com>

Hi Gurus,
For my question below I got a lot of  replies, thank you all for your
replies. I'd like to thank
David Harrington
Michael Schulte
Frank Smith
Crist Clark
Joey Beetle
sreenath sarikonda
I do apologize if I've missed any names, your advises are duly respected.
The answers which hit the mark are from David and Sreenath. The password for
the user account expired. I reset the password and was back in the game.
Thanks a bunch to you Sun Managers.


I've been working on an issue with cron for a particular user and have run
out
of options: I checked if the files perms are ok for the
/var/spool/cron/crontabs/user, the user is root and group is the users
primary
group. I did a HUP on the cron daemon.

I've a user for whom when I run the cron, I get

"user 28893 c Fri Jan  7 11:43:00 2005 ts=9"

The same script runs for other users.

I'd appreciate if you can help me with some pointers.


I'll summarize.

Regards,

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


From omon.edeki at nowellgroup.com  Mon Jan 10 15:43:50 2005
From: omon.edeki at nowellgroup.com (Omon Edeki-NOWELL)
Date: Mon, 10 Jan 2005 14:43:50 -0600
Subject: SUMMARY: Can't Read Disk Label (Thanks Mike!)
References: <49FBA8FADF231D45B5D642C4D4DA0506A8E6A4@tecomsrv095.dic.ezone>
Message-ID: <002801c4f755$17b623b0$210110ac@nowell7irgz46n>

Thanks to Mike David Chanslor for helping me to fix my "Can't Read Disk
Label" error.

There were 2 problems:

1.the system could not recognize the CD-ROM drive so we had to reconfigure
the slave-master jumper settings.

2.the cd-rom drive was faulty. it could not read properly from cd's, so we
put in a a SONY CDROM drive and it ran smoothly. Below is how our
correspondence went (later messages come first):

Omon

#####


Hello! - I work for the power company in Alabama and we have over 300
ultra5/10's...! And I am so ready to upgrade hardware, but in short we
have done a lot of trouble-shooting on these guys. You have the classic
case of IDE blues... What I'm saying is the ultra5/10's ( I keep saying
that because depending on the year most had SAME boards ) are VERY
particular about the jumper settings. We now stick to Seagate IDE drives
because we consistently get the same behavior.
 In a nut shell - you just need to try cable select vs. the MASTER/SLAVE
jumper settings...

If I had the box the first thing I would do is MAKE SURE YOU HAD
OPENBOOT up to date with 3.31
Then I would jumper the IDE drive on IDE bus #1 for MASTER
Next, jumper the CDROM for MASTER on #2 IDE bus.
 - power on and check probe-ide

If that failed, I would try HDD as Primary master on IDE#1 and CDROM as
slave on IDE#1

If that didn't work, I'd think it was a blade and start trying cable
select settings on IDE#1

Mike


-----Original Message-----
From: Omon Edeki-NOWELL [mailto:omon.edeki at nowellgroup.com]
Sent: Friday, January 07, 2005 2:14 PM
To: Chanslor, Michael David
Subject: Re: Can't read disk label.


Dear Mike,

I did a probe-ide and here's what we got from the ok commandline prompt:

ok probe-ide
    Device 0 (Primary Master)
        ATA Model: IBM-DTLA-30.015

    Device 1 (Primary Slave)
        Not Present

    Device 2 (Secondary Master)
        Not Present

    Device 3 (Secondary Slave)
        Not Present

ok

[END]

Is it telling us something about not seeing the cd-drive? Thanks in
advance!

Omon

----- Original Message -----
From: "Chanslor, Michael David" <MDCHANSL at southernco.com>
To: "Omon Edeki" <omon.edeki at nowellgroup.com>
Sent: Friday, January 07, 2005 9:51 AM
Subject: RE: Can't read disk label.


What happens when you power on and do OK> probe-ide
Does it see the disk drive?

Work with a lot of these guys, be glad to help...
Have had similar problems


Mike Chanslor



-----Original Message-----
From: sunmanagers-bounces at sunmanagers.org
[mailto:sunmanagers-bounces at sunmanagers.org] On Behalf Of Omon Edeki
Sent: Thursday, January 06, 2005 1:47 PM
To: sunmanagers at sunmanagers.org
Subject: Can't read disk label.


Dear Sun managers,

Good afternoon. I am trying to install Solaris 8 on a Sun Ultra 10 Sparc

workstation. This system formerly had Solaris 8 on it, but we decided to
do
a reinstallation to configure things the way we wanted them.Before doing

that we encountered a few problems

1.The system failed to recognize the Dell Ultra Sharp Monitor, when the
signal cable was plugged into the port behind the computer. To solve
this we
researched and were told to remove the Video card attached to the mother

board, and this we did.The system now recognized the dell monitor and
booted
in Solarid 8.

2.We ran sys-unconfig to reconfigure the system to change the hostname
on a
reboot.We powered off the computer.Afterward we would go through the
sys-unconfig program in an indefinite loop.So I powered off the computer

each time i got stuck.

3.Lastly I tried to do a STOP A to try to go to the 'ok' prompt in order
to
do a clean installation of Solaris 8 Installation CD, by running

ok boot cdrom [ENTER]

Here's what I am getting:

Boot device: /pci at 1f, 0/pci at 1,1/ide at 3/cdrom at 2,0:f File and args:
Can't read disk label.
Can't open disk label package

Can't open boot device
ok
[end]

I want to do a clean installation of the Solaris 8 CD on the sparc ultra
10
computer.Please can anyone help me out? I looked at the solaris 8
installation guide, and it asked that I follow this procedure. Please
respond if you can help.Thanks a million!

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


From martini at raider.llnl.gov  Mon Jan 10 17:55:53 2005
From: martini at raider.llnl.gov (Dave Martini 1)
Date: Mon, 10 Jan 2005 14:55:53 -0800 (PST)
Subject: SUMMARY: Upgrading Veritas Volume Manager which option?
Message-ID: <200501102255.j0AMtrXE026486@raider.llnl.gov>

Thanks to the many replies.

I don't need a new license key when upgrading from Veritas Volume 
Manager 3.0.4 to 3.5. But, I have to install the new licensing package
VRTSlic.

Everyone suggested I upgrade using the upgrade_start and upgrade_finish
scripts which is what I will do.

Below is my original question:

I'm currently running Veritas Volume Manager 3.0.4 on Solaris 8
> with my root disk encapsulated.
> I want to upgrade to Solaris 9 and Volume Manager 3.5.
> 
> Do I need a new License Key for this upgrade to 3.5 or will
> my 3.0.4 license key work? Is there a charge for this new key?
> 
> Also, the upgrade guide for 3.5 says I have two options to do the
> upgrade.
> 
> Option 1.
> 
> Use the pkgadd command on top of my existing software.
> 
> Option 2.
> 
> Use the upgrade_start and upgrade_finish scripts on the 3.5 CD.
> 
> Has anyone used option 1 before?? Did you have any problems? Option
> 1 does not unencapsulate the root disk and there could be problems
> with the machine re-booting if something goes wrong.
> 
> Thank You.
> Dave Martini
> LLNL
> ___________________________________________
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From Szabadkai.Eva at malev.hu  Tue Jan 11 06:55:41 2005
From: Szabadkai.Eva at malev.hu (=?iso-8859-2?Q?Szil=E1gyin=E9_Szabadkai_=C9va?=)
Date: Tue, 11 Jan 2005 12:55:41 +0100
Subject: Summary: Nawk script
Message-ID: <2498AB062812764899A12174698C387E0E9288@fhbmsg02.malev.hu>

Thanks to all for their prompt response. The best was

    Pal Baranyai       and
    Peter Ondruska

My original question was how  to delete *trc files beeing older then today.
Of course it works with find.
The solution is:

     touch timestamp
     find <dir>  -type f -name \*.trc  !   -newer  <timestamp>   -exec rm -f
{}  \;

Best regards,
Eva
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From sguillemette at metrocastcablevision.com  Tue Jan 11 10:23:26 2005
From: sguillemette at metrocastcablevision.com (Shawn Guillemette)
Date: Tue, 11 Jan 2005 10:23:26 -0500
Subject: SUMMARY-- Re: tape drive questions
In-Reply-To: <D0DCF6B7C413384A9EBBB3B08BCBB42A14C85123@brspm202.exbr01.exch.eds.com>
References: <D0DCF6B7C413384A9EBBB3B08BCBB42A14C85123@brspm202.exbr01.exch.eds.com>
Message-ID: <41E3EF6E.5030709@metrocastcablevision.com>

Thats to all for the suggestions.

I think Im going to take the extra time to sutdown the machine while 
removing the drive.

Thanks you all


Bonfieti Junior, Angelo wrote:

>Shawn,
>there's no problem at all in (dis)connecting an external tape device from
>your box. I've done this thousands of times before.
>Just make sure the tape unit is turned off.
>You can also attach it to any other sun box, if this is your wish, without
>having this box down.
>
>WHen u do back this tape unit to it's original place (e250 solaris7 box), it
>probably will work as if nothing had happened. If u get any kind of trouble,
>just remove the old links in /dev/rmt (I suppose you just have the 0 -zero-
>combination) and perform a dvrconfig;tapes that solaris will recreate the
>new combinations for your tape device.
>
>Hope it helps,
>/angelo
>
>-----Original Message-----
>From: sunmanagers-bounces at sunmanagers.org
>[mailto:sunmanagers-bounces at sunmanagers.org] On Behalf Of Shawn Guillemette
>Sent: Monday, January 03, 2005 7:44 PM
>To: sunmanagers
>Subject: tape drive questions
>
>
>I have a e250 running Solaris 7 with a ds3 scsi tape drive. The e250 is 
>in production and we are looking to use the tape drive else where. 
>Whoever, the drive may have to return to its original owner. ;-)
>
>
>I guess the question is does any one for see any issues with 
>disconnecting this external scsi drive from the e250 while its in 
>production and during office hours?
>
>And what are my chances of adding the device back with out taking the 
>box down?
>
>
>Thanks
>
>  
>

-- 
Shawn Guillemette
MetroCast Cablevision
Network / Operations

There is No Place like 127.0.0.1 
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From guenthere at yahoo.com  Tue Jan 11 11:42:44 2005
From: guenthere at yahoo.com (Ed Guenther)
Date: Tue, 11 Jan 2005 08:42:44 -0800 (PST)
Subject: SUMMARY: flash archives
Message-ID: <20050111164244.89036.qmail@web40908.mail.yahoo.com>

Well I did not get an answer but I know what the
problem actual was.  In my original post I incorrectly
stated that the flar file is copied over, in fact it
is not copied at all.  The problem occurs when you
boot a new, uninstalled client box off the install cd,
manually configure the boot disk with multiple
paritions (/usr, /var, /root, etc.), and then choose
"install by flar".

The partitions created are not all mounted during the
FLAR install.  Only the /root partition is mounted. 
So the install tries to put everything in the root
partition.  So even though I created a /usr filesystem
the contents of it are being shoved into the root
partition under a directory called /usr.

For those who do not know, the v20z and v40z boxes are
i86pc architecture (x86).  Creating a
jumpstart/DHCP/PXE-boot server to install the FLAR was
not tested.

I'm using Solaris 10 beta 72.  Sun is trying to
replicate the problem to fix it.

-Ed.

original post below
I've opened a case with Sun on this problem but they 
don't feel it's urgent. So.... 


I have several new v20z and v40z's here. I installed 
one box with Solaris 10 beta 72. I created a flash 
archive with: 


flarcreate -s -n install /install.flar 


I booted a second v20z (client) and connected it to 
the first using a crossover cable. I booted the 
client from the Sol10 install cd and opted to install 
using a flar image. The system finds the flar master 
server, copies the image but then gives me this error:

  "Could not extract flash archive" and the install 
bombs out. 


I've looked through the archives of this list and 
found one similar case but his solution doesn't solve 
my problem. 


I even feel that I know why this message occurs! The 
root partition must be big enough to hold the entire 
flar. But I do not want the root parition to be 3 
gig! It's wasted space that I need on other 
filesystems. 


I confirmed that this is the problem by allowing the 
system to have three gig in /root. The flar worked 
and the system installed. 


I've contacted sun about this but they feel that hard 
drive space is cheap so why bother? Just waste the 
space and move on. 


Does anyone have a solution so that I can use only 500

meg for the /root partition? This is very urgent. 





		
__________________________________ 
Do you Yahoo!? 
Take Yahoo! Mail with you! Get it on your mobile phone. 
http://mobile.yahoo.com/maildemo 
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From manibs999 at hotmail.com  Tue Jan 11 12:10:00 2005
From: manibs999 at hotmail.com (Mani Savirigana)
Date: Tue, 11 Jan 2005 17:10:00 +0000
Subject: SUMMARY: Any suggestions on Sun Cluster 2.2 issue
Message-ID: <BAY2-F40071AC799F9A2066BF54EC880@phx.gbl>

Hi Managers,

We have Sun cluster 2.2 , Volume manager 3.4, Raidmanager 6.22 and oracle
Cluster nodes are node1(E220R) and node2(E220R).These two nodes are 
connected to two disk arrays(A1000s)

One of the cluster nodes ,node1(primary for oracle) crashed and the oracle 
was switched to node2.
it did not happen automatically. When we are not able to access our in-house 
application, I found that there
is a problem with the cluster node node1. When  I unplumb the logical 
interface for oracle on node1 ,
it switced to node2. Now our application is running fine on node2.

Before that I have added 4 drives to the arrays(Two A1000s , both are 
connected to both the nodes of the cluster)
extended the veritas volume and it's mirror .But the server(node1) crash 
occured after 3-4 days

The errors I found from the logs are all related to snmpxd.

	 	From /var/adm/messages

snmpdx: local pdu process error
snmpdx: agent_process(): bad community from 10.40.33.26

The cpu on node1 paniced

when checked with format on node1 it is in single user mode now

format is able to identify the internal hard drives and also c1t5do and 
c2t5d0. And for the rest of the drives
drive not available :reserved is attached in format listing

When we did format
c1t5d0 No error with this drive
c1t5d1< drive not available: reserved >
c1t5d2< drive not available: reserved >
c1t5d3< drive not available: reserved >
c1t5d4< drive not available: reserved >
c1t5d5< drive not available: reserved >
c2t5d0 No error with thsi drive
c2t5d1< drive not available: reserved >
c2t5d2< drive not available: reserved >
c2t5d3< drive not available: reserved >
c2t5d4< drive not available: reserved >

And raid sense key errors related to battery expiration. All the batteries 
were expired on the A1000 arrays.I also
have to replace the batteries on these A1000 arrays.

Can I simply rejoin node1 with the cluster ?
Can you please suggest me  any resolution quickly.

Thanks,
Mani

_________________________________________________________________
Make team work really work! Work together, stay connected! 
http://www.microsoft.com/india/office/experience/  With Microsoft Office 
System.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From ruhnke at us.ibm.com  Tue Jan 11 16:04:37 2005
From: ruhnke at us.ibm.com (Chris Ruhnke)
Date: Tue, 11 Jan 2005 15:04:37 -0600
Subject: SUMMARY:  "unknown priority name *" in syslog.conf
Message-ID: <OF2D01E595.97D9B292-ON86256F86.0072B7FD-86256F86.00736D55@us.ibm.com>

... and the winners are: Jason Santos <jason.santos at megaslow.net> and 
Ric Anderson <RIC at Opus1.COM> 

Who both educated me (better than the man page did!) that

local2.debug    /var/adm/messages

will cause ALL messages to be logged ("debug" and higher priorities, 
"debug" is the lowest priority).

and that is what I wanted to do!


Thanks guys!!!
--CHRis

Chris H. Ruhnke
Technical Services Professional
IBM Global Services
Dallas, TX


O'Toole's Law:  Murphy is an optimist.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From oetiker at ee.ethz.ch  Tue Jan 11 17:12:51 2005
From: oetiker at ee.ethz.ch (Tobias Oetiker)
Date: Tue, 11 Jan 2005 23:12:51 +0100 (MET)
Subject: [Summary] multihomed host and anti-spoofing filters
Message-ID: <Pine.GSO.4.61.0501112307200.13101@abin.rr.rgum.pu>

My Question:

> We have a multi-homed Solaris box serving as a boot-server in
> several subnets. The machine does not route.
>
> Our Networking People now want to introduce anti-spoofing filters
> on each of the subnets routers/switches.
>
> This raises an interesting problem.
>
> When a client host opens a connection to the multi-homed servers
> main interface (which is not in the local subnet) the answer will
> be sent through the servers interface connected to the clients
> subnet.
>
> This does not play well with the anti-spoofing filters (or so our
> network people tell us)
>
> Is there any way to tell a Solaris box to always answer on the
> same interface as it received the packet in the first place ?

Thanks to Philipp Buehler, Michael Horton, Matthew Stier, Darren
Dunham, Crist Clark for providing their insights.

This is what I have learned:

a) At the networking level, there is no association between any
   "incoming" and "reply" packets.  That could only be done at the
   application layer (it's not generally done).  The interface from
   which a packet leaves is determined by its destination alone.

b) Use ipfilter's NAT rules to do some source routing.  If a packet
   has a particular (solaris) source address, force it to leave via
   a particular interface.

   http://www.sunmanagers.org/pipermail/summaries/2002-May/001645.html

c) One solution is to prevent the packets from going to the wrong
   interface in the first place by makeing sure that the hosts file
   has the 'main name' for the multihomed-box on all its addresses,
   and have all the clients refer to the host via its 'main name'.
   Machines will then automatically use the closest interface and
   thus forego any problems.

   192.168.1.1   host0   host
   192.168.2.1   host1   host
   192.168.3.1   host2   host
   192.168.4.1   host3   host

d) disable anti-spoofing on the switch ports of that machine ;-)

cheers
tobi

-- 
 ______    __   _
/_  __/_  / /  (_) Oetiker @ ISG.EE, ETL F24.2, ETH, CH-8092 Zurich
 / // _ \/ _ \/ /  System Manager, Time Lord, Coder, Designer, Coach
/_/ \.__/_.__/_/   http://people.ee.ethz.ch/oetiker +41(0)44-632-5286
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From martini at raider.llnl.gov  Tue Jan 11 18:48:59 2005
From: martini at raider.llnl.gov (Dave Martini 1)
Date: Tue, 11 Jan 2005 15:48:59 -0800 (PST)
Subject: SUMMARY: Drive compatibility
Message-ID: <200501112348.j0BNmxXE027933@raider.llnl.gov>

Thanks to all the replies!

The Ultra 10 is IDE and so it won't work in the Blade 1000
which is SCSI. I have an Ultra 60 machine which I believe
is SCSI which I'll try in the Blade 1000. I also have a 280R 
which is deffinitely SCSI.

Dave Martini
LLNL

My original question is below

Does anyone know if a CD rom drive in an Ultra 10
will work in a SunBlade 1000? What about one from an Ultra 60?
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From santhakumar at bigpond.com  Tue Jan 11 19:29:19 2005
From: santhakumar at bigpond.com (Siva Singam Santhakumar)
Date: Wed, 12 Jan 2005 11:29:19 +1100
Subject: Summary: Seagate LTO on 420r
In-Reply-To: <200501112209.OAA28711@redwood.taos.com>
References: <200501112209.OAA28711@redwood.taos.com>
Message-ID: <41E46F5F.10608@bigpond.com>

Dear managers,

Thanks to Darren for pointing me to the drivers.  I didn't any 
documentation with me and I used only the SUN download center documents.

Thanks to the other admins for sending me their solutions.



Darren Dunham wrote:

>>Dear Managers,
>>I am trying to attach a Seagate Viper 200 Ultrium LTO to a 420r running 
>>Sol 8.  probe-scsi-all can see the tape drive but not the OS. Tape is 
>>attached to the LVD on PCI1 slot. What am I missing?
>>    
>>
>
>One of the new Sun adpaters does not have drivers in the base OS.  You
>have to install them afterward.  (The qus drivers).  The documentation
>with the adapter should tell you if this is necessary.
>
>If it's visible to probe-scsi-all but not to the OS, I would suspect
>that the adapter drivers are missing.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From sebastien.daubigne at atosorigin.com  Wed Jan 12 04:27:39 2005
From: sebastien.daubigne at atosorigin.com (Sebastien Daubigne)
Date: Wed, 12 Jan 2005 10:27:39 +0100
Subject: SUMMARY: Changing T3+ IP address without rebooting ?
In-Reply-To: <006b01c4f7fc$884a81a0$326dbba3@daubigne>
Message-ID: <000901c4f888$f6e85b00$326dbba3@daubigne>

Thanks to Paul, Michael, Chris, Dustin and David.

Everyone agrees that I have to reboot the T3 for the new IP to take effect
(is this a Microsoft-like product ?).
David pertinently pointed out that the "set" command manual talks about
setting "boot parameters", something I didn't notice
but that makes sense.

--
Sebastien DAUBIGNE
Sebastien.daubigne at atosorigin.com
mailto:Sebastien.daubigne at atosorigin.com>  - +33(0)5.57.26.56.36
AtosOrigin Infogerance - IMS/ERP/Pessac


-----Message d'origine-----
De : sunmanagers-bounces at sunmanagers.org
[mailto:sunmanagers-bounces at sunmanagers.org]De la part de Sebastien
Daubigne
Envoyi : mardi 11 janvier 2005 17:42
@ : sunmanagers at sunmanagers.org
Objet : Changing T3+ IP address without rebooting ?


Hi,

[ This is some kind of follow-up of a previous message posted during
christmas week : my last try before opening a support call, in case last
message has been missed by people away ]

Here is the problem : We moved 2 T3+ arrays from one network to another.
I connected to them via console port, disabled RARP, and *statically* set
IP/Netmask/Gateway (using the "set" command).

Now I'm trying to ping the *new* IP :
- From the T3+ : I get a "sendto error 5033", which is an error from the IP
stack, not a "no response" !
- From a server on the same subnet : I get no response.

However, when I try to ping the *old* IP *from* the T3+,  the T3+ responds
OK !
Note that pinging another IP from the same subnet as the *OLD* IP gives no
error (just a "no response").

It seems the T3+ did not take into account the new IP, even if the "set"
command reports the new IP, as if the old IP was still stored in the
internal IP stack !
As the 2 T3+ have different firmware revisions, I guess it's not a bug, and
I suspect the T3+ needs to be rebooted to take the new IP.

Any idea ?

Here are some traces :

sfret3p1:/:<68>set
bootmode  auto
bootdelay 3
sn        105458
ip        55.8.132.22
netmask   255.255.255.192
gateway   55.8.132.1
tftphost  0.0.0.0
tftpfile
hostname  sfret3p1
timezone
vendor    0301
model     501-5710-02(51)
revision  0200
logto     *
loglevel  3
rarp      off
mac       00:20:f2:00:43:cf

===> NEW address : Ping ERROR
sfret3p1:/:<67>ping 55.8.132.1
ping (55.8.132.1): 56 data bytes
ping: sendto error 5033
ping: wrote 55.8.132.1 64 chars, ret=-1

===> RANDOM address : Ping ERROR
sfret3p1:/:<71>ping 82.66.29.56
ping (82.66.29.56): 56 data bytes
ping: sendto error 5033
ping: wrote 82.66.29.56 64 chars, ret=-1

===> OLD address : Ping OK
sfret3p1:/:<66>ping 143.165.4.16
ping (143.165.4.16): 56 data bytes
143.165.4.16 is alive

===> Another address on the same subnet as the OLD address : Ping KO
sfret3p1:/:<70>ping 143.165.4.17
ping (143.165.4.17): 56 data bytes
no answer from 143.165.4.17



--
Sebastien DAUBIGNE
Sebastien.daubigne at atosorigin.com - +33(0)5.57.26.56.36
AtosOrigin Infogerance - IMS/ERP/Pessac
_______________________________________________
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 paca at sci.fi  Wed Jan 12 05:20:47 2005
From: paca at sci.fi (Petri Asikainen)
Date: Wed, 12 Jan 2005 12:20:47 +0200
Subject: SUMMARY: Why processes are swapping on solaris 8
In-Reply-To: <41E4915D.7040106@sci.fi>
References: <41E4915D.7040106@sci.fi>
Message-ID: <41E4F9FF.1080001@sci.fi>

Thanks to all, here's quick summary of messages from
Petri Kallberg , John Leadeham and Joohyun Cha
Given information leads me check postgresql database configuration and 
usage.

Cheers,

Petri

Petri Asikainen wrote:
> Apache 2.0.52  (prefork) is swapping, every forked new process seems to 
> reserve swapspace, but top and vmstat are showing that theres free 
> memory available. Server is wery slow when it get multiple hits...
> 
> 
 >Can I  use any specific gcc options when compiling apache to avoid 
swapping?


No. Swapping is controlled by OS. I guess you could lock some of your
pages in memory but I don't really think is going to solve your problem.
(and anyway it would require extensive modifications to application)


> Why os allocating swapspace even there's free memory?
If it is using (as opposed to reserving) swap then
there is another cause. Perhaps postgres is doing
full table scans?


> #vmstat -p
>      memory           page          executable      anonymous filesystem
>    swap  free  re  mf  fr  de  sr  epi  epo  epf  api  apo  apf  fpi
>  5659632 1722640 160 73 21  0   0    8    0    0    0    0    0   85

The first line of vmtstat is a meaningless average since booting.
If it is not the first line, there is no sign of a problem.

> 
> fpo  fpf
> 21   21
> 
> 
> # swap -s
> total: 249432k bytes allocated + 192736k reserved = 442168k used, 
> 5386720k available

Reserving swap space is just a mapping within process's address map not
physical action. (ie. hard disk access)
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From PolachakJM at NSWC.NAVY.MIL  Wed Jan 12 07:02:09 2005
From: PolachakJM at NSWC.NAVY.MIL (Polachak Jason M CONT DLVA)
Date: Wed, 12 Jan 2005 07:02:09 -0500
Subject: SUMMARY: Solaris 10 will not mount cdrom
Message-ID: <BDD35658B859D51187AB00508BB820300633C778@nswcdlvaex12.nswc.navy.mil>

-----Original Message-----
From: Polachak Jason M CONT DLVA 
Sent: Wednesday, January 12, 2005 6:57 AM
To: 'sunmanagers at sunmanager.org'
Subject: SUMMARY: Solaris 10 will not mount cdrom



Simply login as root and type "svcadm enable smserver"

Thanks to all of you who responded so quickly

-----Original Message-----
From: Polachak Jason M CONT DLVA 
Sent: Tuesday, January 11, 2005 1:55 PM
To: 'sunmanagers at sunmanagers.org'
Subject: Solaris 10 will not mount cdrom


My problem is that I installed Solaris 10 on my new V440 server and it will
not mount any cdroms.  I have tried inserting a "Sun" cd then
cd cdrom   then  ls  and there is nothing.  Then I look down and the server
has ejected the cd.  This happens no matter what cd I put in there.  The
server works fine.  Its just that after install I am now trying to accredit
the machine and it will not accept a cdrom to copy the accreditation
software onto it.  I have also stopped and restared volume manager.  Does
anyone know an answer?  

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


From broschi at id.ethz.ch  Wed Jan 12 10:24:11 2005
From: broschi at id.ethz.ch (Erwin Broschinski)
Date: Wed, 12 Jan 2005 16:24:11 +0100 (MET)
Subject: SUMMARY: Disk >  137.44GB not supported in Blade 100/Solaris 9?
In-Reply-To: <XFMail.20050112135637.broschi@id.ethz.ch>
Message-ID: <XFMail.20050112162411.broschi@id.ethz.ch>

Thanks to the answers and hints from:
Michael Horton, Ryan Krenzischek, Julio Carrasco, Ric Anderson, Tim Chipman,


"IIRC, the Blade-100 still uses an ATA-4 controller which
maxs out somewhere around 127GB at the hardware level." (Ric Anderson)

This seems to be the same limit as for a Ultra 5 according to this summary:
29 Nov 2004 14:41 Miriam von Zuben SUMMARY: Big disk in Ultra 5


May I suggest that this fact is being added to this list's  FAQ:
Subject: 5.11) * How do I use an EIDE/ATA disk larger than 8GB?

Erwin


On 12-Jan-2005 Erwin Broschinski wrote:
| Hi
| 
| I have mounted a 160GB Disk in a Blade 100 with Solaris 9 and only see this:
| 
| iostat -E:
| 
| dad1     Soft Errors: 0 Hard Errors: 0 Transport Errors: 0 
| Model: SAMSUNG SP1614N  Revision: TM100-30 Serial No: S016J10XC08135   
| Size: 137.44GB <137438952960 bytes>
| 
| format:
| 
| 1. c0t2d0 <SAMSUNG SP1614N cyl 255 alt 2 hd 16 sec 255>
|           /pci at 1f,0/ide at d/dad at 2,0
| 
| Part      Tag    Flag     Cylinders       Size            Blocks
|   0 unassigned    wm       0              0         (0/0/0)         0
|   1 unassigned    wm       0              0         (0/0/0)         0
|   2     backup    wu       0 - 254      508.01MB    (255/0/0) 1040400
| 
| I found the following solution to make at least some of the disk visible:
| 
| erased the disk label:
| dd if=/dev/zero of=/dev/rdsk/c0t2d0s2 bs=1b count=16
| 
| and entered 65535 cylinders in format:
| 
| format> type
| 
| 
| AVAILABLE DRIVE TYPES:
|         0. ST315310A
|         1. SAMSUNG SP1614N
|         2. other
| Specify disk type (enter its number)[1]: 2
| Enter number of data cylinders: 
| Enter number of data cylinders: 65535
| Enter number of alternate cylinders[2]: 
| Enter number of physical cylinders[65537]: 
| Enter number of heads: 
| Enter number of heads: 16
| Enter number of data sectors/track: 255
| Enter rpm of drive[3600]: 7200
| Enter format time[default]: 
| Enter cylinder skew[default]: 
| Enter track skew[default]: 
| Enter tracks per zone[default]: 
| Enter alternate tracks[default]: 
| Enter alternate sectors[default]: 
| Enter cache control[default]: 
| Enter prefetch threshold[default]: 
| Enter minimum prefetch[default]: 
| Enter maximum prefetch[default]: 
| Enter disk type name (remember quotes): "SAMSUNG SP1614N"
| selecting c0t2d0
| [disk formatted, no defect list found]
| format> pa   
| 
| PARTITION MENU:
|         0      - change `0' partition
|         1      - change `1' partition
|         2      - change `2' partition
|         3      - change `3' partition
|         4      - change `4' partition
|         5      - change `5' partition
|         6      - change `6' partition
|         7      - change `7' partition
|         select - select a predefined table
|         modify - modify a predefined partition table
|         name   - name the current table
|         print  - display the current table
|         label  - write partition map and label to the disk
|         !<cmd> - execute <cmd>, then return
|         quit
| partition> pr
| Current partition table (SAMSUNG SP1614N):
| Total disk cylinders available: 65535 + 2 (reserved cylinders)
| 
| Part      Tag    Flag     Cylinders         Size            Blocks
|   0       root    wm       0 -    64      129.49MB    (65/0/0)       265200
|   1       swap    wu      65 -   129      129.49MB    (65/0/0)       265200
|   2     backup    wu       0 - 65534      127.50GB    (65535/0/0) 267382800
|   3 unassigned    wm       0                0         (0/0/0)             0
|   4 unassigned    wm       0                0         (0/0/0)             0
|   5 unassigned    wm       0                0         (0/0/0)             0
|   6        usr    wm     130 - 65534      127.25GB    (65405/0/0) 266852400
|   7 unassigned    wm       0                0         (0/0/0)             0
| 
| 
| At least 127.5 GB for a 160 GB disk :^(
| 
| Is this currently the upper limit?
| 
| 
| Cheerz
| Erwin
| 
| 
| 
| 
| 
|                                                          ''`'
| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~O-O~~~~~~~
| Erwin Broschinski               Tel:    +41 1 632 4281
| Swiss Fed. Inst. of Technology  Fax:    +41 1 632 1022 
| ETH Zentrum CLU B2              E-Mail: broschi at id.ethz.ch
| 8092 Zurich                      
| Switzerland                     
| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| 
| "Ceterum censeo, 'Parvam Mollim' esse delendam."  (nach Cicero)
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From cbar44 at tsg.cbot.com  Wed Jan 12 13:24:33 2005
From: cbar44 at tsg.cbot.com (Christopher L. Barnard)
Date: Wed, 12 Jan 2005 12:24:33 -0600 (CST)
Subject: SUMMARY: Using an SDS raid5 partition
In-Reply-To: <Pine.GSO.4.58.0501120832470.2248@shalmaneser.tsg.cbot.com>
References: <Pine.GSO.4.58.0501120832470.2248@shalmaneser.tsg.cbot.com>
Message-ID: <Pine.GSO.4.58.0501121223550.2248@shalmaneser.tsg.cbot.com>

I asked:

> I am configuring a new system with SDS/LVM/Fred/whatever-Sun-is-calling-
> their-disk-volume-manager-this-week.  I have configured mirrors and stripes
> many times, but this is my first RAID volume.  I have successfully metainit
> d5:
>
> d5 -r /dev/dsk/c2t8d0s5 /dev/dsk/c2t9d0s5 /dev/dsk/c2t10d0s5 \
>       /dev/dsk/c2t11d0s5 /dev/dsk/c2t12d0s5
>
> I know that with mirrors of UFS filesystems it is not necessary to newfs
> the /dev/md/dsk partition but with stripes it is.  Guessing that a RAID
> partition is similar, I attempted to newfs /dev/md/dsk/d5.  No joy.
> What is the magic needed to mount and start using the RAID partition?
>
> Thanks, and I will summarize.

The answer:

chill.  don't be so impatient.  Go out and have a cup of coffee.  Thats a
lot of disk space that SDS has to configure.  Wait until "metastat d5" says
ok before trying to newfs it.  And yes, I was trying to newfs the raw
partition, that was just a typo.

Thanks to all who answered so fast.

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


From sun1sol at yahoo.com  Wed Jan 12 17:47:48 2005
From: sun1sol at yahoo.com (ert weerr)
Date: Wed, 12 Jan 2005 14:47:48 -0800 (PST)
Subject: SUMMARY: what's chewing up the swap space?
In-Reply-To: <41E306C5.6@genesyslab.com>
Message-ID: <20050112224749.66721.qmail@web52203.mail.yahoo.com>

>Is that possible to check what applications are 
>sitting in the swap and not releasing the used 
>swap area? 
>We have an old box what we shouldn't reboot, but a
>badly written application always chewing up the free 
>swap space.
>We need to identify this application and kill 
>it if it's possible.

I received a lot of useful solutions for my original
question.
Thanks to everyone who spent time to reply for my
email!

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

You can do /usr/ucb/ps -alxww and then check out the 
processes that have a large SZ.

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

ps -ef or prstat  will show the amount of memory (real

& virtual) used by a process.  if the memory used by a

process grows other time, it is  your memory-leaker.

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

ps -elf shows a SZ column which is the size of the 
process in pages. 
Multiple that by the output of page size to find the 
total size of the processes in bytes.  pmap -S <pid> 
will show the swap reservation of a process if you're 
running under Solaris 9.

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

prstat -s size

also check for files in (or applications writing logs
into) /tmp

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

"ps -el" should show a large SZ for the offending 
process (remember SZ is in pages, so its a lot bigger 
than it looks).  Solaris 8 has "ps -eyl" which shows
SZ
in kilobytes.  Once you find a large process, you can 
use tools in /usr/proc/bin to learn more about it,
or just watch it for a while to see if it grows.

Also run lsof on /tmp.  It could be some process is 
writing big files there.  Since /tmp is a memory file 
system, that can chew up a ton of swap space.

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

"lsof" can be used to view open files in tmp/swap with
a link count of 0 (zero).

Thanks!

John


		
__________________________________ 
Do you Yahoo!? 
All your favorites on one personal page  Try My Yahoo!
http://my.yahoo.com 
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From mot1 at btinternet.com  Thu Jan 13 06:18:55 2005
From: mot1 at btinternet.com (MOT ABU)
Date: Thu, 13 Jan 2005 11:18:55 +0000 (GMT)
Subject: Summary ./install: Permission denied
Message-ID: <20050113111855.81753.qmail@web86209.mail.ukl.yahoo.com>

I used the anon=0 option.
All is well now.
Cheers.
 
 
> I have exported a cdrom as an nfs.
> Mounted the cdrom.
> As root I get
> #./install
> bash: ./install: Permission denied
> 
> What am I doing wrong?
> 
> Thanks.


		
---------------------------------
 ALL-NEW Yahoo! Messenger - all new features - even more fun!  
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From dan.cave at gmail.com  Thu Jan 13 07:21:41 2005
From: dan.cave at gmail.com (Dan Cave)
Date: Thu, 13 Jan 2005 12:21:41 +0000
Subject: SUMMARY: Booting jumpstart client to use gigabit interface.
In-Reply-To: <E671455BAF0BFC43B8B31A6455348E7F06FA0766@srvmeth10.sis.co.uk>
References: <E671455BAF0BFC43B8B31A6455348E7F06FA0766@srvmeth10.sis.co.uk>
Message-ID: <2731bd3b050113042156030de2@mail.gmail.com>

Thanks to Jeremy Ahl for his v.prompt reply below.

at the OK prompt, do a show-nets

This will list the available interfaces and allow you to cut-and-paste into
a new devalias for "net"
Then "boot net" will work the correct interface
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From avriette at gmail.com  Thu Jan 13 11:08:52 2005
From: avriette at gmail.com (Alex J. Avriette)
Date: Thu, 13 Jan 2005 11:08:52 -0500
Subject: SUMMARY: rules.ok being ignored by jumpstart client
In-Reply-To: <f866414505011307555577b39f@mail.gmail.com>
References: <1b56057d05011305567f400feb@mail.gmail.com>
	<f866414505011307555577b39f@mail.gmail.com>
Message-ID: <1b56057d050113080871fb0488@mail.gmail.com>

So the list of people who responded with help:

Casper Dik
Dale Hirchert
Conor Svensson
Matt Clausen
Lai Wei
Peter van Genmert
Fernando Boveda

First and foremost, Dale was very helpful. After going through just
about everything in the book, he suggested from a Sunsolve document
that I remove the hostname and ip_address directives from the sysidcfg
file. I did so, without any success. Around the same time, both
Fernando and Peter suggested I check the netmasks of the client and
server.

After setting the netmask on the server from /16 to /24, to match the
client's /24, the jumpstart proceeded in an automated fashion, which
is as it should be.

This frustrates me, as all the mounts worked, and the rules were
plainly visible to the client. The sysidcfg was correct and crafted
per the manpage, and Sun's Jumpstart Technology Blueprint book (I've
been using jumpstart for a long time, but this is a new install at a
new network). So it strikes me that everything was configured that it
*could* be working correctly, but that it simply chose not to.

Dale is correct in saying that the hostname and the IP address are
configured during boot, and having them in the sysidcfg file is not
necessary, however, it seems to me that it should be possible to have
a machine come up with one ip, supplied by the bootserver, and to
install the OS with the configuration of another ip. Afterall, the
installed server does not use the bootserver to ascertain its
configuration information. In this case, specifying the hostname and
ip address would be required.

But I digress.

I will say that I am once again impressed by the fabulous volume of
responses I received from Sunmanagers, and most especially so by the
diligence and indeed correctness on the part of Dale, Peter, and
Fernando (which is not to say I do not appreciate any of the rest of
it. I'll take anything after days of pounding on something that
"should work").

Cheers,
alex

I have included the original message below for the spiders.

I have a peculiar jumpstart problem. I've included output of the
client jumpstarting, here:

Rebooting with command: boot net - install
Boot device: /pci at 1f,700000/network at 2  File and args: - install
1000 Mbps FDX Link up
Timeout waiting for ARP/RARP packet
Timeout waiting for ARP/RARP packet
29200 1000 Mbps FDX Link up
SunOS Release 5.9 Version Generic_117171-07 64-bit
Copyright 1983-2003 Sun Microsystems, Inc.  All rights reserved.
Use is subject to license terms.
whoami: no domain name
Hardware watchdog enabled
SC unretrieved msg JAN 13 06:09:06 2005 UTC [Host System has Reset]
Configuring /dev and /devices
Using RPC Bootparams for network configuration information.
Skipping interface bge3
Skipping interface bge2
Skipping interface bge1
Configured interface bge0
Searching for configuration file(s)...
Using sysid configuration file
10.2.3.251:/export/jumpstart/config/dc-mvncrmas2/sysidcfg
Search complete.
The system is coming up. Please wait.
Begin system identification...
Starting remote procedure call (RPC) services: sysidns done.
System identification complete.
Generating software table of contents [this may take a few minutes...]
Table of contents complete.
Starting Solaris installation program...
Searching for JumpStart directory...
not found
Warning: Could not find matching rule in rules.ok
Press the return key for an interactive Solaris install program...^C
Executing JumpStart preinstall phase...
Searching for SolStart directory...
Checking rules.ok file...
^Cawk: trying to access field -2
record number 1
/sbin/suninstall: test: argument expected
# ^C
# df -h
Filesystem             size   used  avail capacity  Mounted on
dc-suntools:/export/jumpstart/install/Solaris_9/Tools/Boot
                       20G   1.1G    18G     6%    /
swap                   1.9G   560K   1.9G     1%    /tmp
proc                     0K     0K     0K     0%    /proc
mnttab                   0K     0K     0K     0%    /etc/mnttab
/tmp/devices           1.9G   560K   1.9G     1%    /devices
/tmp/dev               1.9G   560K   1.9G     1%    /dev
10.2.3.251:/export/jumpstart/install
                       20G   1.1G    18G     6%    /cdrom
fd                       0K     0K     0K     0%    /dev/fd
10.2.3.251:/export/jumpstart/config/dc-mvncrmas2
                       20G   1.1G    18G     6%    /tmp/sysid_config.141
/usr/sbin/install.d/install_config
                       20G   1.1G    18G     6%    /tmp/_install_config
# cd /cdrom
# ls rules.ok
rules.ok
# ls -la
total 202
drwxr-xr-x   5 root     other        512 Jan 12 20:31 .
drwxr-xr-x  17 root     other        512 Dec 29 16:45 ..
-r--r--r--   1 root     root          85 Jun 17  2004 .cdtoc
drwxr-xr-x   5 root     other        512 Dec 27 07:21 .install
lrwxrwxrwx   1 root     other         32 Dec 27 06:59 .install_config
-> ./Solaris_9/Misc/.install_config
-r--r--r--   1 root     root         462 Dec 27 07:08 .slicemapfile
-r--r--r--   1 root     root          20 Jun 17  2004 .volume.inf
-r--r--r--   1 root     other         22 Dec 27 07:21 .volume.inf.2
-r--r--r--   1 root     root        7052 Jun 17  2004 Copyright
drwxr-xr-x   2 root     other        512 Jan 12 15:28 Profiles
drwxr-xr-x   8 root     root         512 Dec 27 07:20 Solaris_9
-rw-r--r--   1 root     other        377 Dec 30 04:23 actual_size
-rwxr-xr-x   1 root     other        383 Jan 12 20:31 addclient.sh
-rwxr-xr-x   1 root     other      55475 Dec 29 15:43 check
-rwxr-xr-x   1 root     other        257 Dec 27 07:21 installer
-rw-r--r--   1 root     other       4808 Jan 12 15:27 rules
-rw-r--r--   1 root     other         80 Jan 12 16:17 rules.ok
# cat rules.ok
any      -            -       Profiles/mvno.profile -
# version=2 checksum=3435
# ls -la Profiles/mvno.profile
-rw-r--r--   1 root     other        492 Jan 12 15:28 Profiles/mvno.profile

And so on. On the server, we see:

# showmount -a
dc-mvncrmas2:/export/jumpstart/install
dc-mvncrmas2:/export/jumpstart/install/Configs/dc-mvncrmas2
dc-mvncrmas2:/export/jumpstart/config/dc-mvncrmas2
dc-mvncrmas2:/export/jumpstart/install/Solaris_9/Tools/Boot
# grep dc-mvncrmas2 /etc/bootp*
/etc/bootparams:dc-mvncrmas2
root=dc-suntools:/export/jumpstart/install/Solaris_9/Tools/Boot
install=10.2.3.251:/export/jumpstart/install boottype=:in
sysid_config=10.2.3.251:/export/jumpstart/config/dc-mvncrmas2
install_config=10.2.3.251:/export/jumpstart/install
rootopts=:rsize=32768

It is extremely frustrating, as you can see. The directories all get
mounted, sysidcfg is seen and used, the client box is configured
properly, and for some reason, although rules.ok is sitting right
where it is supposed to be, it is being ignored. I can then proceed
manually through the install, but this means that all the filesystem
configuration I've done in my profiles, and all the users/patches/etc
that get done by finish scripts are going to be done by hand. And I've
got a lot of these to do.

When it says it's searching for the "jumpstart directory", watching
snoop seems to indicate that it is looking for the 'install' parameter
from bootparams. However, that parameter is clearly defined, and even
mounted on the system (/cdrom).

Thanks in advance for any help.

alex

--
Alex Avriette
avriette at gmail.com

-- 
Alex Avriette
avriette at gmail.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From glowe at sbcglobal.net  Thu Jan 13 11:29:37 2005
From: glowe at sbcglobal.net (Grant Lowe)
Date: Thu, 13 Jan 2005 08:29:37 -0800 (PST)
Subject: SUMMARY: IBM 7337 tape library
Message-ID: <20050113162937.38109.qmail@web81707.mail.yahoo.com>

Well, I got no answers from anybody, but I got the problem resolved.  It turns out that the IBM 7337 tape library is actually a Breece Hill Q2.15 tape library.  I went through the front panel of the library, and changed the ID of the library from an IBM to Q2 (the Fleece Hill ID), and now NBU is happy.
 
Original question:
 
Hi gurus.
 
I'm running an E420, Solaris 9, fully patched, with 2 GB RAM and 2 procs.  I would like to know if anybody has experience with this tape library.  I'm having a very frustrating time with this and getting it to work with NetBackup.  robtest can move tapes around, but for some reason I'm getting some errors that say "Unable to initialize robot"...so I can't label tapes, let alone use them for backup.  Any help is appreciated.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From sreeredhat at yahoo.com  Wed Jan 12 13:23:52 2005
From: sreeredhat at yahoo.com (sreenath sarikonda)
Date: Wed, 12 Jan 2005 10:23:52 -0800 (PST)
Subject: SUMMARY: Serial console session not working. 
Message-ID: <20050112182352.82511.qmail@web54210.mail.yahoo.com>

Thanks to     
   Daniel Vega, Chris.CTR.Schroen@ faa.gov,  Peter
Ondruska, Santhan Perampalam, Ed Guenther,  Harvey
Wamboldt , Casper.Dik at Sun.COM , John Christian and
Chuck.

Here is the response which most of them gave.
set
output-device=ttya
input-device=ttya
and reboot the machine. I cannot reboot machine so
left it alone.

OR just leave of fthe kbd/mouse cable and it will
start sending data to ttya .This is what happened to
most of my machines.However I do not have kbd/mouse (I
use terminal emulator)attached to any of them.Yet they
started using ttya as default. However this particular
machine missed.

 diag-out-console parameter should be set to true when
using with the RSC and false for use with ttya. I am
not using RSC so this doesn't apply to me.

Cheers,
Sree.
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From doug at trackmaster.com  Thu Jan 13 11:43:55 2005
From: doug at trackmaster.com (Doug Hubbard)
Date: Thu, 13 Jan 2005 08:43:55 -0800
Subject: OT: Apache Auth help needed Summary
Message-ID: <41E6A54B.9070503@trackmaster.com>

 Thanks to Steve Rieger and Yura Pismerov.
This is an issue with the crypt library being used (old version) The
solution is to choose a different encryption method for the passwords
(MD5 for example).
Original Question

  I am configuring an Apache server (Apache 2.0.49) with mod_auth_mysql
  and the server
  is only requiring the first 8 characters of the password (it behaves
  the
  same way with mod_auth_dbm).
  I would like to force it to require the full 20, but can't seem to
  find
  anything on this topic.
  Just to give an example,
  If I set my PW to 1longtestpassword
  I can log in with PW of 1longtes
  Any and all help appreciated. --

--

*******************
Doug Hubbard - IT Manager
TrackMaster, an Equibase Company
email doug at trackmaster.com
Website www.trackmaster.com
*******************

This message is intended only for the use of the Addressee and may
contain information that is PRIVILEGED and CONFIDENTIAL. If you are not
the intended recipient, you are hereby notified that any dissemination of
this communication is strictly prohibited. If you have received this
communication in error, please erase all copies of the message and its
attachments and notify us immediately.
Thank you!

[demime 1.01b removed an attachment of type text/x-vcard which had a name of doug.vcf]
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From chadalavada2002 at yahoo.com  Thu Jan 13 17:13:18 2005
From: chadalavada2002 at yahoo.com (Srini Chadalavada)
Date: Thu, 13 Jan 2005 14:13:18 -0800 (PST)
Subject: Summary diff btw df -k & du -sk
Message-ID: <20050113221318.64605.qmail@web40429.mail.yahoo.com>

Hi Managers,
  Thank-you all for response. yesterday evening we
took downtime and restarted the database that fixed
the problem. It is better to install lsof to check
which process is using that file. I still need to
install that utility .
Thanks,
Srini

The DBA may have to restart the database. The deleted
file space will not return unitl the database, or
whatever process that has it open, closes the file.
One way to determine which process still has it open
is to use 'lsof'. You can download it from
www.sunfreeware.com. There are lots of options but a
quick-n-dirty w/o RTM is 'lsof|grep -yourfile-'.

Regards,
--
Reggie Beavers



		
__________________________________ 
Do you Yahoo!? 
All your favorites on one personal page  Try My Yahoo!
http://my.yahoo.com 
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From martini at raider.llnl.gov  Thu Jan 13 19:28:36 2005
From: martini at raider.llnl.gov (Dave Martini 1)
Date: Thu, 13 Jan 2005 16:28:36 -0800 (PST)
Subject: SUMMARY: Veritas Upgrade Problem
Message-ID: <200501140028.j0E0SaXE001007@raider.llnl.gov>

I ended up talking to a SUN support rep to solve this.

vxconfid wasn't starting at bootup because I didn't have 
any disks in rootdg. Once I added 1 disk then vxconfigd started
each time I rebooted the machine.

I ran a vxprint -hft and noticed that my volumes showed up
but were DISABLED.

So, I ran a 

# vxdg import delphi
and it said that the disk group was already imported. ( this I believe
was handled by the uprade_finish script)

I then did a 

# vxrecover -g diskgroupname -sb

and that ENABLED my volumes in vxprint.

At that point I was set.

I also renamed the /VXVM3.5-UPGRADE directory so that I would not
get the WARNING: upgrade did not finish error messages each time
the system rebooted.

Below is my original question:

I'm upgrading from Volume Manager 3.0.4 to 3.5 and from Solaris
8 to 9.

I followed the procedures in the VM 3.5 install guide and did the
upgrade_start script. It changed my encapsulated root disk back
to the normal /dev/dsk file system and that went ok. I then removed
my old volume manager packages and then I upgraded
to Solaris 9 and that went fine. I then added the VM 3.5 packages
and I then ran the upgrade_finish script.

When I reboot my system I get several messages

NOTICE: vxvm: upgrade was not complete
NOTICE: vxvm: upgrade is not complete
VXVM provider configuration daemon is not accessable.

I was expecting the upgrade_finish script to convert my root
disk back to encapsulation but it didn't do this. Is this normal?

I also expected my VM volumes on my external disk pack to function
but I can't mount them. The upgrade_start script commented these
out in my /etc/vfstab file but the upgrade_finish script did not
uncomment them out.

I also notice that vxconfid does not start on bootup I have to type vxconfigd
to start it.

I can start vea ok but no volumes are listed.

Are there extra steps I need to complete after the upgrade_finish script?

Thanks.
Dave Martini
LLNL
_______________________
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From neil at quiogue.com  Thu Jan 13 21:09:52 2005
From: neil at quiogue.com (Neil Quiogue)
Date: Fri, 14 Jan 2005 10:09:52 +0800
Subject: SUMMARY: Issues with T3: Battery and Unmountable Volume
In-Reply-To: <1442031469.20041110111032@quiogue.com>
References: <1442031469.20041110111032@quiogue.com>
Message-ID: <683531527.20050114100952@quiogue.com>

Since I was just able to finish the fixing of the "shadow" volume
recently, this is a long overdue summary:
1. You can use the T3extender.tar.Z can be obtained from Sun to fix
this.  Alternatively, you can use the following commands:

  telnet into T3 as root
  fru stat    (list battery status)
  .bat -n u1pcu1    
  .id write busage u1pcu1 0      (reset battery one)

  .bat -n u1pcu2
  .id write busage u1pcu2 0      (reset battery two)

  refresh -s         (next scheduled refresh, also gives battery status)

2. Wrong position is as it says, it's in the wrong position so just
swapping the disks properly would fix this.
3. For the shadow volume, it was fixed with 'boot -w' but this was
made possible because the T3 was not in production.  Otherwise, one
would need to connect the T3 back to the partner group then unmount it
before removing it.

The negative effect of 'boot -w' is that you would need to set
everything again since it removes volumes, etc.  That means you would
need a console for this.

The following people helped with this:
Govind Kulkarni (T3extender) and Paul Roetman (for the command summary
in the battery).
Mike Busse for the shadow volume.

Thanks again.

Regards,

Neil Quiogue

Wednesday, November 10, 2004, 11:10:32 AM, you wrote:

NQ> Hello,

NQ>       I just have a few questions on Storedge T3:
NQ>       1. The Power Supply was already replaced and yet when I do an
NQ>       fru stat, it still sees the battery as fault.  I tried the
NQ>       refreshing and changing busage of u1pcu1 to no avail.  The
NQ>       syslog shows that it's not ok also.
NQ>       2. What does it mean that the disk is in the wrong position?  I
NQ>       can't seem to find out any distinguishing mark on where each
NQ>       disk would be placed.
NQ>       3. I keep seeing a 'u2d1-d9' volume yet there's no such disks.  I
NQ>       tried unmount and removing it but it does not work.  Somehow the
NQ>       config sees it as such and I can unmount it properly:

NQ>       Here's what I see:
NQ>              <> vol unmount vol1
NQ>              <> vol remove vol1
NQ>              ERROR (200008) VN_MOUNTED: volume already mounted
      
NQ>       Any pointers would be much appreciated.

NQ> Regards,

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


From ors.tiszay at ericsson.com  Fri Jan 14 07:17:09 2005
From: ors.tiszay at ericsson.com (=?iso-8859-1?Q?=D6rs_Tiszay_O_=28LA/ETH=29?=)
Date: Fri, 14 Jan 2005 13:17:09 +0100
Subject: SUMMARY: IOWAIT on a slice with no filesystem?
Message-ID: <F005CD411D18D3119C8F00508B0874800D72FE5D@ehubunt100.eth.ericsson.se>

OK, the answer was fairly quick and easy: I was looking at the wrong
place. Since the root disks are encapsulated, there is no point in
looking at the Solaris volumes, Veritas does not access the disks via
the c0t0d0sx interface. It is also useless to use iostat is such cases.
Answers are quoted below. Thanks to:

NO UCE
Sebastian Daubigne

for their quick and accurate answers

=========================================================================
You have the OS mirrored via VxVM. The root disk c0t0d0 is encapsulated.
Let's look at WHAT VXVM does exactly:

Part      Tag    Flag     Cylinders         Size            Blocks
  0       root    wm   20624 - 21712        1.50GB    (1089/0/0)   3146121
  1       swap    wu   21713 - 24615        4.00GB    (2903/0/0)   8386767
  2     backup    wu       0 - 24619       33.92GB    (24620/0/0) 71127180
  3          -    wu       0 -     0        1.41MB    (1/0/0)         2889
  4          -    wu       1 - 24619       33.91GB    (24619/0/0) 71124291
  5 unassigned    wm    5083 - 20623       21.41GB    (15541/0/0) 44897949
  6        var    wm       1 -  1452        2.00GB    (1452/0/0)   4194828
  7        usr    wm    1453 -  5082        5.00GB    (3630/0/0)  10487070


In this layout, VxVM created Slice 3 as the private region and Slice 4 as
the data region.

When the OS tallks to any of these slice, VxVM talks to slice 4 at offset
??? to get all data. That is Not slice 0,1,5,6 or 7. As such you see the
overlap in slices. VxVM has it's own view of the disk once encapsulated and
does not use the normal S0,S1 ...

As such your I/O.

Maybe use vxstat to see what is REALLY going on, where the io is..

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

FS Partitions are not directly accessed when you have the rootdisk under
VxVM control.
Usually slice 4 is the the partition for the public region of VxVM.
The public region holds the data, so it is normal to see I/O activity on
this particular partition.

Now if you want to track the FS that drives the I/O activity, you can use
'vxstat' (e.g. "vxstat -v -i 5 -c 5").
=========================================================================


And the original post: 

I have a Netra T1405 with 4 CPU, 4GB of RAM and 4GB swap running Solaris 8
(kernel Generic_108528-18). The machine runs Oracle OID (Oracle 9.0.1.4.0,
OID 3.0.1.1.0) which stores its data on a T3. Oracle is put under Veritas
Cluster (3.4) control.
>From time to time, without any recognisable time or usage patterns the
machine experiences very high iowait which can only be stopped by restarting
the Oracle resource group. What makes it interesting for me is that iostat
(see printout below) seemes to indicate high disk busy rate on a slice of
the mirrored root disks which does not even have a file system on it, and of
course it's not the swap slice either.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From cbar44 at tsg.cbot.com  Fri Jan 14 10:45:55 2005
From: cbar44 at tsg.cbot.com (Christopher L. Barnard)
Date: Fri, 14 Jan 2005 09:45:55 -0600 (CST)
Subject: SUMMARY: mount at boot without error RAID partition
In-Reply-To: <Pine.GSO.4.58.0501131457240.2248@shalmaneser.tsg.cbot.com>
References: <Pine.GSO.4.58.0501131457240.2248@shalmaneser.tsg.cbot.com>
Message-ID: <Pine.GSO.4.58.0501140945210.2248@shalmaneser.tsg.cbot.com>

I asked:

> I now have a V210 (Solaris 9) up and running with a 278 gig raid5 partition.
> (thanks guys).  Unfortunately, it is still not working right.  Whenever
> this box is rebooted, it fails to fsck this massive partition and the
> server drops into single user mode.  I manually fsck it (it takes abt 10
> minutes), there are no errors, and I can mount it and continue.  I've
> tried several things which required several reboots.  At the last reboot
> I got tired of fsck-ing a huge partition when I know there will be no
> problems, so I mounted it manually without a fsck first.  It worked
> just fine.  I checked the Usual Suspects, including the /etc/vfstab and
> in particular the "device to fsck" column, and everything is fine.  I
> really do not want to have to manually mount this filesystem every time
> the server reboots.  Does anyone have a suggestion on getting this
> partition to mount without bogus errors on bootup?

The Solution:

I guess I am just not used to RAID5 devices.  A disk had gone bad, hence
the error on bootup.  But with one bad disk a RAID5 can still be used, so it
would come up if I manually mounted it.  And yes, I have logging enabled on
all partitions on all my servers, including this one.  My understanding is
that logging doesn't mean you never have to fsck again, it just greatly
decreases the possibility that you might have to.

Thanks to:

rossman at columbia.edu
olli <olli at hydrodevil.com>
Alan Pae <alan.pae at gmail.com>
<David.Harrington.ctr at dla.mil>
Neil Quiogue <neil at quiogue.com>
Darren Dunham <ddunham at taos.com>
Tim Chipman <chipman at ecopiabio.com>
James Noyes <jnoyes-sml at retrogeeks.com>
Jason Santos <jason.santos at megaslow.net>
Michael Horton <Michael.Horton at acntv.com>

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


From Shahla.Tabarzadi at mail.house.gov  Fri Jan 14 11:49:27 2005
From: Shahla.Tabarzadi at mail.house.gov (Tabarzadi, Shahla)
Date: Fri, 14 Jan 2005 11:49:27 -0500
Subject: SUMMARY: Failure to compile tcp wrappers on Solaris 8
Message-ID: <E76621C896A74A45B22D82676BB6606D0856345F@hrm34.houseexchange.us.house.gov>

Look at this one:
# /usr/ccs/bin/make REAL_DAEMON_DIR=/usr/sbin sunos5

cc -O -DFACILITY=LOG_AUTH -DHOSTS_ACCESS -DPARANOID -DNETGROUP
-DGETPEERNAME_Bc
sh: cc: not found

*** Error code 1

make: Fatal error: Command failed for target `tcpd.o'

Current working directory /opt/tcp_wrappers_7.6-ipv6.4

*** Error code 1

make: Fatal error: Command failed for target `sunos5'



________________________________

Shahla Tabarzadi 
US House of Representatives 
Network Configuration and Engineering Branch 
202.226.6266 (voice and fax)


From gdseas at yahoo.com  Sat Jan 15 19:31:03 2005
From: gdseas at yahoo.com (GDS)
Date: Sat, 15 Jan 2005 16:31:03 -0800 (PST)
Subject: SUMMARY: Network attached storage ideas
Message-ID: <20050116003103.73018.qmail@web52803.mail.yahoo.com>

Dear Managers,

Thank you for your responses.

Let me start by clarifying that the issue with Solaris
9 and NFS is not with Solaris but with how the Snaps
interact with Solaris 9 NFS. Sorry for the confusion.
On to the responses.

It seems that (money notwithstanding)  the safest bet
is with Network Appliance. There have also been
numerous alternatives proposed (will have to visit all
the vendors) along with Sun Storage and Apple Xserve
RAID.  

As foar as I know, for my customer, the Sun solutions
are prohibitively expensive. The Apple Xserve RAID may
be a nice robust system but customer does not want
another OS in the  current environment. I have to look
at prices for NetAPPs and the other options. Thank you
again for all your help.  Special thanks to Tim
Chipman for his pointers and detailed response.

Following you will find the original question and 
responses (in no particular order).


===================== ORIGINAL QUESTION
=====================
Long-time listener, first time caller.
We have been using Snap servers for quite some time to
perform our backups (using our home-grown scripts).
They have been OK but flaky but we always managed with
some hacks to deal with the problems.
However, we just moved to Solaris 9 and NFS mounting
is a nightmare.  I am looking into network attached
storage ideas for something in the order of 2+TB.
RAID is good, performance requirements are not strict
(we will only use them for backups).
I looked at www.fastora.com and I like their IDE-type
of setup.
Does anyone have any suggestions or recommendations
for what products or vendors to look for?

===================== RESPONSES =====================
From:	Fergason.com  
Network Appliance.  Might be a bit more expensive that
you want to 
spend, but they are true appliances.  Incredibly easy
to setup and maintain.
They have a Nearstore 200, which starts at 8Tb and
expands to 96Tb.

From:	"Grzegorz Bakalarski"
Anyway you may look for e.g. following vendors:
www.proware.com.tw
www.infortrend.com
www.nexsan.com,
www.synetic.net
www.acnc.com
www.maxtronic.com,
www.fibrenetix.co.uk and many, many similar.
proware, synetic, acns(JetStor), maxtor(Arena) use 
more or less the same raid controller (intel based)
and interface,
Infortrend is build around PowerPC chipset (similar to
SUN Storage
3310/3311/3510 array). NexSan uses own chipset base on
motorola...

From:	"Linux Admin"
it really depends on how deep your pockets are. Netapp
R200 is nice,
NetAPP OS is awesome. The unit is a cadilac as far
pricing. it plays
nice with Sun, Linux , MS.


From:	"David Foster" 
We are using Nexsan ATAboy2 arrays (2.6TB) and have
been
extremely happy with them. Cheaper because they use
commodity ATA drives, but performance has been similar
to our Sun T3 arrays. Just be sure to configure a
hot-spare
as these drives are slightly less reliable than the
much
more expensive SCSI or FC disks. We indeed had one go
bad,
the unit picked up the hot-spare and didn't miss a
beat,
and Nexsan replaced the drive immediately.

The web-based interface for the Nexsans is superb.


From:	"John Malick"
Sun's 5310 NAS storage is great. Check it out.



From: 	"Tim Chipman"
-not quite clear why NFS is a nightmare for you ..
maybe if you want to 
clarify what you were trying / what you are trying to
mount - backup 
etc .. further clarification/resolution might be
possible there (?)

-for what it is worth: We've got an IDE_based disk
array from AC&NC, 
"JetStor III", which I've been very happy with. Price
was good (~$9k 
USD for 8 x 200 gig, or ~1.4tb in raid5 without
hot-spare) and similar 
things can be had for slightly less per gig from them
especially if 
larger capacity is your target (12 / 14 / 16 disk
arrays ...).  For 
what 
it is worth, AVOID like the plague and large Maxtor
IDE disks, we had 
serious problems with them.. Seagate disks have better
warranty and 
fewer failure issues.

-we also recently bought an array from Winchester
Systems, also IDE 
based, an "older refurb" model that I got specifically
b/c it was cheap 
and we needed another brick for a development/test
box. This one I 
haven't tested much yet, but seems to work well, so I
can also 
recommend 
them. IF you can get an off-lease or "last years unit"
you will find 
great prices. OR go the full splurge and buy the
latest and greatest, 
whatever you prefer :-)

anyhow, in general, for sure IDE based units will be
WAY! cheaper than 
SCSI/Fcal disk models. Performance will be a bit lower
too but probably 
not a big issue for your needs. also, for ref, our
Jetstor unit with 
200gig IDE disks is about the same performance as an
A1000 (all scsi 
disks) so clearly the IDE-based units aren't "junk"
(as some might tell 
you?)

finally, as you have found, there are MANY vendors
selling this sort of 
stuff ... (IDE disk arrays) ... lots of options. Just
be picky with the 
brand of disk used, IMHO Maxtor is toxic to be
avoided. (we had bad 
blocks pile up on 6 of our 8 disks in the array in
their first 18 
months 
of service, totally unacceptable).

From:	Send an Instant Message "Solo" 
Hi there

This may at first sound like a joke, but I am dead
serious when I say 
you ought to check out Apple's Xserve RAID. They're
beating anything 
else hands down both in price and performance. Even
Oracle were amazed 
by it and they now run Apple internally for all their
SAN infra.

The Xserve RAID uses IDE disks and you connect to it
through 
fibrechannel. Scales up to 5.6 TB and has a number of
certifications 
for various SAN products including Veritas Volume
Manager.

http://www.apple.com/xserve/raid

From:	"Alex J. Avriette"
Hi Gregor.

It's tough to beat the Apple XServe RAID for
dollars-per-gig. In a
similar homegrown setup a couple years ago, I
installed 2tb of XServe
and it did the trick just fine.


From:	"Derek Olsen" 


 I dont understand your statement about nfs mounting
being a nightmare 
on solaris 9 but I will give you some info I came
across while looking 
for NAS storage besides netapp and snap appliance.

  This product is something I considered in an
environment where I 
could 
use ide drives but I dont have any experience with it
beyond the 
documentation.   

http://www.open-e.com/index.php?g=product&#9001;=de&aid=153


   The EMC NS line of NAS devices when populated with
ata drives 
instead 
of fibre channel comes in at a pretty reasonable
price.     Obviously 
what I consider a reasonable price was in relation to
the importance of 
the project I was working on and the associated
revenue so it may not 
be 
right for you. 

From:	jastrologo

netapp fas250.




		
__________________________________ 
Do you Yahoo!? 
Yahoo! Mail - Helps protect you from nasty viruses. 
http://promotions.yahoo.com/new_mail
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From zoo11 at hst.co.kr  Mon Jan 17 01:46:00 2005
From: zoo11 at hst.co.kr (Joohyun Cha)
Date: Mon, 17 Jan 2005 15:46:00 +0900
Subject: SUMMARY: Message Queue full and App. fail
Message-ID: <20050117064600.GB19308@mail.hst.co.kr>

Hi. Managers.

Thanks for you quick and kind advice especially..
  Paul Wilkinson
  John Leadeham

My original post is at the end of message.

Although I couldn't find exact answer yet, one thing I can sure is that
there is no OS's bug about message queue. Below is a part of Paul's messages.

> Sun are pretty open about OS bug's, search http://sunsolve.sun.com
>
> Message queues have been around in Unix for so long that I'd have
> thought the code involved is well tested and stable, I'd be surprised
> if the OS is at fault here. I've programmed apps using message
> queues without such problems.

I pretty agree with him. I will try to find any bug of application more 
carefully not the OS's.

Thanks all.

I asked.
--- 
Hi. Managers

I've problem with my servers. These servers have E3500 with 8 CPUs (Sol 7)
and SF4800 with 8 CPUs (Sol 8). Several Application (not common application
such as RDBMS, Web, WAS etc. They are all developed by our programers) is
running on them. All the Application program is consist of several independant
processes and using several message queue to IPC.
The symptom is...
 1) all of sudden some message queue is filled up fullly and msgsnd() of
    sending process failed. Thus Application program failed. I noticed
    that using 'ipcs'
 2) system is slow down. CPU's %sys is over 50%. by sar, vmstat (normal
    system load is %sys < 20)
 3) msgrcv() of receive process is returned very slowly. msgrcv() is called
    with IPC_WAIT.

Same problem occurs on systems
 1) is running on over 200days
 2) across on nealy all platform and OS version. (E3500 with Sol 7 and SF4800
    with Sol 8)
 4) has no system log messages. (/var/adm/messages)
 5) has process using message queue

The problem is gone with system reboot. I don't know what reason occurs this.
I want to know If Application is assumed properly programed, is there any
kernel bug related to this issue? I wonder why this problem is gone with
system reboot Do I need to reboot the server periodically to prevent this
problem? Is there any recommandation about system reboot from Sun?

Please, May I take your bright ideas?
Thanks in advances.
-- 
------------------------------------------------------------------------------
From:
main(){int a=122,j=11;while(a>-50){a=a>0?a:111;printf("%c",a);a=j==49?46:a-j;
j*=j;j=j%2?j:7;j*=j%8?1:0;}printf("%c1\n",49);}
------------------------------------------------------------------------------
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From rob.de.langhe at belgacom.be  Mon Jan 17 05:54:58 2005
From: rob.de.langhe at belgacom.be (rob.de.langhe at belgacom.be)
Date: Mon, 17 Jan 2005 11:54:58 +0100
Subject: SUMMARY: VxVM DRL combined with UFS-logging: filesystem corruption?
Message-ID: <20E53B61AAB3994EA15681023EE52418D19E27@AE0007.BGC.NET>

Thx to Juergen Rothenanger for a swift reply:

SUN's Alert Notification 57636 (in SunSolve, search on document-ID
57636) describes on the spot what happened.

Altough this document suggests to disable the UFS-logging by putting
explicitly the monut-option "nologgin" in "/etc/vfstab" and leave VxVM
perform its logging via the DRL, we will prefer the other strategy:
Not encapsulating the root-disk(s) in VxVM, but stick with Solstice
Disksuite (or Solaris Volume Manager, for Solaris-9 and up) to mirror
the boot-disks. Any other less critical disks and partitions can still
be managed with VxVM because (currently) more easy to use.

Still, for any VxVM-managed (logical) partitions, by default they are
created by VxVM with logging (DRL) enabled. So we will install the
documented patches so that VxVM will add the "nologging" option in
"/etc/vfstab" to avoid the logging-conflict with DRL.

Regards,

Rob

-----Original Message-----
From: sunmanagers-bounces at sunmanagers.org
[mailto:sunmanagers-bounces at sunmanagers.org] On Behalf Of
rob.de.langhe at belgacom.be
Sent: 17 January 2005 10:57
To: sunmanagers at sunmanagers.org
Subject: VxVM DRL combined with UFS-logging: filesystem corruption?

Hi,

last week we had a system with "/" filesystem corrupted so badle, that
it had to be restored from backup tape.

Surely we do not want this to happen again, so after careful discussion
with SUN about what could have caused such bad filesystem corruption,
SUN noticed that both UFS-logging was activated via the options in
"/etc/vfstab", and the Veritas Volume Manager's "Dirty-Region-Logging"
(DRL) was enabled at the same time.
This was a non-supported situation according to SUN, but they failed to
present me clear documented reasons that such combination indeed is the
root-cause for a filesystem corruption.

Can anyone point me to documented behaviour with these 2 logging-methods
activated at the same time? I admit, this situation doesn't sound too
healthy, but I inherited this environment since a few weeks and want to
show good reasons why it should be 'corrected'.

Thx a lot in advance !

Rob


**** DISCLAIMER ****
http://www.belgacom.be/maildisclaimer
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


**** DISCLAIMER ****
http://www.belgacom.be/maildisclaimer
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From Conor.Svensson at evolution.net  Mon Jan 17 08:13:58 2005
From: Conor.Svensson at evolution.net (Conor Svensson)
Date: Mon, 17 Jan 2005 13:13:58 -0000
Subject: Summary: Solaris & Veritas SAN question
Message-ID: <9920848EF398D311BDC400508BF339F90290AD69@ldnisp14.evolution.net>

Thanks to the following members of this list for their prompt replies.

 

Jon Hudson

Jeff Barratt-McCartney

Ddelija

 

I have pasted the body of Jon Hudson's comprehensive reply below.

 

 

In a well tuned SAN on local multimode fiber (62.5, <150m) all scsi requests
should be processed (ECT;End completion Time) in under 500ms. Should be able
to get down to 100ms often, but under 500ms is the target. Sometimes really
low, 

 

Once you go to single mode (long haul) you start having scsi time out
issues. You limitation ends up being between 120-220miles depending on
equipment. This is just due to physics limitation of fibre. Now, you can of
course have devices in there that allow multi-hop synchronous replication,
but it's only for replication. 

 

Now in a test lab I saw a Nishan box in CA allow hosts to mount disk in NY.
But the Nishan had to buffer and cache everything so they could deal with
all the timeout issues. I can't remember what scsi times out at, I'd have to
check. But most solutions I've seen are under 120miles.

 

As far a comparing internal drives to FC disk array, that really depends on
what the internal drives are. Fiber channel drives do really well either
internal or external as long as you are dealing with good size data sets in
that can be carved up into neat 2048 frames. Basic scsi can out run fiber
channel when dealing with lots of really small files. FC ends up having to
pad the 2048 frame causing addition interrupts on the hba. I feel pretty
safe saying in most cases a 2Gb attached array tuned properly with 10k
drives will out perform any internal or scsi attached scsi array. But I
could construct special cases with really small files where this isn't true.
You know how it works, you can always construct a niche test that will prove
your point =)

 

So for example an v880 with internal FC drives can perform well since it's
fiber, but now as well as an external array because it's FCAL

(loop) and not FCSW (switched).

 

But with the right equipment you "should" be able to do several kilometers
and be ok, but I'm not sure of the performance numbers. We

(Finisar) make long haul single mode SFPs that can do 80Km without a
repeater that might work. But you are probably best off getting Brocade and
Nishan (now own by McData) to show you their solutions. Don't talk to Cisco,
their FC stuff sucks.

 

Make sure you spend the time to properly tune the credit buffers, this can
make or kill long haul stuff.

 

 

Thanks,

 

Conor

 

 

-----Original Message-----
From: Conor Svensson 
Sent: Friday, January 14, 2005 3:51 PM
To: 'sunmanagers at sunmanagers.org'
Subject: Solaris & Veritas SAN question

 

Hi, 

 

We are looking to implement a large scale SPARC Solaris system which has two
main priorities. 

1.	Disk latency - ie fastest possible return of any calls to disk IO 
2.	Failover to remove site 

 

I am investigating using a Veritas to manage a SAN mirror of Fibre Channel
disk arrays. The application writes relatively little data. 

 

Does anyone know how attaching a FC mirror several kilometres away will
affect the disk i/o latency?

How does the i/o latency of an internal disk compare to that of a tuned FC
disk array? 

 

I will summarise, thanks in advance,

 

Conor

 

___________________________________________________________
Evolution is the investment banking and venture capital industry's 
first choice for practical advice on strategy, business process and
the application of advanced technology. 

 


Conor Svensson                         Direct: +44 (0) 20 7898 0348
Evolution                                   Tel:      +44 (0) 20 7664 6640
Peninsular House                       Fax:     +44 (0) 20 7664 6641 
30-36 Monument Street
London EC3R 8LJ                      Email:
<mailto:Conor.Svensson at evolution.net> Conor.Svensson at evolution.net
United Kingdom                         URL:  <http://www.evolution.net/>
http://www.evolution.net
___________________________________________________________
 
The information in this Internet e-mail is confidential and is 
intended solely for the addressee.  Access, copying or re-use 
of information in it by anyone else is unauthorised.  Any views
or opinions presented are solely those of the author and do
not necessarily represent those of Evolution or any of its
affiliates.  If you are not the intended recipient please contact 
Evolution, London, +44 (0) 20 7664 6640
___________________________________________________________
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From mot1 at btinternet.com  Mon Jan 17 08:26:20 2005
From: mot1 at btinternet.com (MOT ABU)
Date: Mon, 17 Jan 2005 13:26:20 +0000 (GMT)
Subject: Summary:Solaris9 scripts which collect system info
Message-ID: <20050117132620.1976.qmail@web86210.mail.ukl.yahoo.com>

Sun's Explorer Data Collector is the one.
 
 
 
Thanks to everyone for swift replies.
Too many to mention.
Some of which include
John Christian
Angelos Kanellopoulos
KrisT
Joon
Rob


		
---------------------------------
 ALL-NEW Yahoo! Messenger - all new features - even more fun!  
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From alex at indo.com  Mon Jan 17 21:14:05 2005
From: alex at indo.com (Alexander Surbakti)
Date: Tue, 18 Jan 2005 09:14:05 +0700
Subject: SUMMARY:How to call a script function when mail server receive
 an email
In-Reply-To: <1103778691.14810.20.camel@loki.amok.local>
References: <41CA2FE8.2000304@indo.com>
	<1103778691.14810.20.camel@loki.amok.local>
Message-ID: <41EC70ED.9000906@indo.com>

dear all,

several days ago i sent this topic to this mailing list and i just 
solved it yesterday night.
most of what i have done is like what disastermaster was told. i have a 
perl file named callscript.pl that needed to be run when our mail server 
get an email from a certain address, at this case the address i use is 
alex at localhost. all the step that i have done is:

shell>vi /etc/aliases
   #make some editing
     alex :    "|/etc/smrsh/callsript.pl"      
   #this is what disastermaster forgot to tell me that the perl file 
must be placed at /etc/smrsh
shell>newaliases

and then i write mail to alex at localhost.  when i check at 
/var/spool/mail_log i found that stat=sent.  it  verified with the  
increament of rows number in my  mysql table  since the callscript.pl  
written  with insert data  function.

earlier the callscript.pl i placed at /var/www/cgi-bin which caused 
stat=service unavailable.

have a nice day,
cheers,
-alex-


From hirdejit at gmail.com  Tue Jan 18 01:21:04 2005
From: hirdejit at gmail.com (Hirdejit Singh)
Date: Tue, 18 Jan 2005 11:51:04 +0530
Subject: SUMMARY: Veritas Volume Removal
Message-ID: <b50ff83405011722211fe4d47d@mail.gmail.com>

The Volume was being used by the Oracle database for temp file
storing. Bringing down the database helped slove the issue.





Original Post 
Hirdejit Singh  to sunmanagers 
  

Hi,

I am unable to remove a volume in veritas.  When removing the same with command

vxedit -g <dg> -rf rm volume-name
gives error that Volume is open or in use and cannot be destroyed.

Even vxdg destroy dg-name gives the same error

Vxprint -ht looks likes this

dg post_mdm     default      default  21000    1105739927.5588.spbill1
dm disk01       -            -        -        -        REMOVED
dm disk02       -            -        -        -        REMOVED
dm disk03       -            -        -        -        REMOVED
dm disk04       -            -        -        -        REMOVED
dm disk05       -            -        -        -        REMOVED
dm post_md03    c6t4d0s2     sliced   2888     71124291 -
v  post_mdm_vol raid5        DISABLED NEEDSYNC 284495872 RAID     -
pl post_mdm_vol-01 post_mdm_vol DISABLED ACTIVE 0       RAID      5/32     RW
sd disk02-01    post_mdm_vol-01 disk02 0       71124291 0/0       -        RMOV
sd disk01-01    post_mdm_vol-01 disk01 0       71124291 1/0       -        RMOV
sd disk03-01    post_mdm_vol-01 disk03 0       71124291 2/0       -        RMOV
sd disk04-01    post_mdm_vol-01 disk04 0       71124291 3/0       -        RMOV
sd disk05-01    post_mdm_vol-01 disk05 0       71124291 4/0       -        RMOV

How do I go about this

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


From rmonahan at ritman.com  Tue Jan 18 03:04:31 2005
From: rmonahan at ritman.com (Richard Monahan)
Date: Tue, 18 Jan 2005 03:04:31 -0500
Subject: SUMMARY:  Netra T1, serial console unable to use input device]
Message-ID: <000a01c4fd34$58339a00$1300a8c0@RITMAN2003>

http://sunportal.sunmanagers.org/pipermail/summaries/2002-July/002041.ht
ml
 
I have the exact same problem as described originally at the bottom of
the URL above. The only difference is that I have a real cisco console
cable as I used to install cisco switches and routers for a while. I can
see all the output from the terminal window, but it will not accept any
input whatsoever. My server is also trying to boot from the net, and I
have no facilities to work with the situation. I wish to change the
boot-device to cdrom and re-install the operating system.
 
Is there a solution on the list? I could not find it. I was able to dump
to the ok prompt when I had keyboard input, but now I am totally stuck
because the only input left on my v100 is the cdrom, but I have no way
to issue a command to boot from it. PLEASE, if anyone can help me I
would be eternally grateful, and promise to learn about posting messages
correctly on message boards(I know that I am supposed to post in the
forums and not email directly to the people, but I am so stuck that I
just wanted to make sure that my cry for help was announced. 
 
If anyone can provide the solution, as well as the correct way to post
my own requests, I promise to follow protocol from this day forth. Thank
you very much.
 
 
 
   - Richard Monahan


From WestB at LondonHydro.com  Tue Jan 18 09:06:33 2005
From: WestB at LondonHydro.com (West, Brian)
Date: Tue, 18 Jan 2005 09:06:33 -0500
Subject: SUMMARY-filesystem on 6120 san is slow
Message-ID: <B6C14F69AE0C60499F9AF872B640D67B68106C@MAILCLUSTER-01.londonhydro.com>

Apparently there was a problem with the firmware version 3.1.2 that is
fixed in newer versions.

The workaround was to run this command on the 6120:
lun  perm  all_lun  rw  all_wwn
Brian.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From feveritt at cisco.com  Tue Jan 18 11:17:26 2005
From: feveritt at cisco.com (Frank Everitt)
Date: Tue, 18 Jan 2005 11:17:26 -0500
Subject: Summary: How To Define An Apple Xserve RAID
Message-ID: <71B88670-696C-11D9-9216-000A959FF96C@cisco.com>

The resolution isn't what I wanted to hear, but this is from Sun. This 
is a known limitation (2tb limit) on solaris 9 for LUNS that present 
themselves to format as physical luns (I.E. hardware type raid and 
certain other companies' arrays).  According to the source code which 
causes this message to print:

>> Comments:
>>
>>
>> from usr/src/cmd/format/ctlr_scsi.c:
>> ==============================================================
>> 2482                 /*
>> 2483                  * this requires us to use SBC-2 commands, which 
>> we
>> 2484                  * don't support
>> 2485                  */
>> 2486                 err_print("The capacity of this LUN is too 
>> large.\n");
>> 2487                 err_print("Reconfigure this LUN so that it is < 
>> 2TB.\n");
>> ==============================================================
>>
>> this shows that the 'format' utility does not use SBC-2 (sbc stands 
>> for scsi block commands) commands.
>> therefore, we are limited to a maximum lun size of 2TB.
>>
>> however, don't confuse this with maximum filesystem sizes.
>> by using SVM or Veritas Volume Manager, you CAN create larger devices
>> on which you CAN put larger filesystems.  however, a hardware device,
>> such as this lun that the customer is trying to configure, can only 
>> be up to 2TB in size.
>> ----------------------------------------------------------------------
>>
>>
>> Sun also said that this is a fairly common industry wide limit that 
>> is just now being broken. It isn't an OS limit but rather a 
>> Storage/SCSI limit. He sent the announcement below as an example.  
>> Looks like we are stuck at the 2 tb limit at this point though.  You 
>> can also get a ton more info on this if you google the keywords "2TB 
>> scsi limitation."


Sun also said that this won't be a problem with Solaris 10's zfs 
filesystem. According to Sun, you put disks and luns in pools and your 
files system will grow and shrink as needed with no intervention and no 
need to grow filesystems.  This works with everything except the OS 
filesystems. I haven't played with any version of Solaris that has zfs 
included, but will be anxiously waiting for release 10.odd_number.....

Thanks to all that replied. For the time being, I'll just need to break 
up the array into a couple of LUNs of a size less then 2TB and move on.


Frank

------------------------------------------------------------
           |             |             Frank Everitt
           |             |             Systems Administrator
         :|||:         :|||:           7025 Kit Creek Rd.
        :|||||:       :|||||:          RTP, NC 27709
     ..:|||||||:.....:|||||||:....     Ph. (919) 392-8885
                         			  FAX. (469) 574 5042
			CISCO SYSTEMS 			  Cell: (919) 624-6098
------------------------------------------------------------
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From sloewenthal at gemini.edu  Tue Jan 18 13:34:20 2005
From: sloewenthal at gemini.edu (LOEWENTHAL Simon)
Date: Tue, 18 Jan 2005 15:34:20 -0300
Subject: SUMMARY: Serial console set-up on U60.
Message-ID: <DOEBKMNALLLEDCCBEFPNEEDMDHAA.sloewenthal@gemini.edu>

Dear all once more,

The solution, thanks to NO UCE was this URL

http://sunsolve.sun.com/search/document.do?assetkey=1-9-72300-1&searchclause
=serial%20mouse

A comprehensive list of OBP commands (OpenBoot 3.x Command Reference Manual)
was sent to me from from Ronny Martin,
http://docs.sun.com/app/docs/doc/805-4436?q=openboot

Thank-you to all who replied as your emails pointed me in the right
direction,
NO UCE
Richard Skelton
Clive McAdam
Ronny Martin
Coy Hile
ben

Regards,
Simon.

---
ISG/Gemini South/AURA
+56.51.205610
www.gemini.edu

-----Original Message-----

Sent: 18 January 2005 05:31
To: Sun Mangers
Subject: Serial console set-up on U60.


Dear all,

I have a U60 with a serial cable connected.  I know that the cable is good
as I tested it on an E250 and U10 and I can get the console prompt via our
Perle CS9000 terminal server.  When I put this into the U60 I get nowt.

Is there any additional configuration I need to do afore I can use the
serial port for console access?

Sorry for this noob question, but its been awhile and Google isn't helping
much.

Regards,
Simon.
--
No virus found in this outgoing message.
Checked by AVG Anti-Virus.
Version: 7.0.300 / Virus Database: 265.6.13 - Release Date: 16/01/2005
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From andy.ford at telindus.co.uk  Wed Jan 19 06:13:25 2005
From: andy.ford at telindus.co.uk (Andy Ford)
Date: Wed, 19 Jan 2005 11:13:25 +0000
Subject: SUMMARY: System logging in batches
In-Reply-To: <1106132610.2043.7.camel@evo.telindus.co.uk>
References: <1106132610.2043.7.camel@evo.telindus.co.uk>
Message-ID: <1106133205.2042.12.camel@evo.telindus.co.uk>

ok ok - everyone is allowed a stupid question once in a while.
I was looking in the wrong place.

The suggestions were to use S|=1, and S|=1 works btw.

Something I use extensively in other scripts.

Thanks

Andy

On Wed, 2005-01-19 at 11:03 +0000, Andy Ford wrote:
> Hi everyone
> 
> I'm not sure whether this is a particular issue related to what I am
> doing ... here goes...
> 
> I am running a perl process that connects to the sysV kernel queues
> (aside). I want this process to log all STDOUT & STDERR stuff to one
> file, so I do this ...
> 
> perl_proc > /var/log/some_log_file.log 2>&1 &
> 
> This does work, however, the STDOUT part logs to the file in batches.
> Can this be changed to log every time the script prints to STDOUT.
> 
> I have played with metalog under gentoo linux and know it's possible.
> 
> I'm not even sure which process controls logging under Solaris 2.8!
> 
> Thanks
> 
> Andy
> --
> perl -e "print qq^bIG VeRN ! ^^qq^#'#Yv#=<D+ ^"
> 
> This e-mail is private and may be confidential and is for the intended
> recipient only.  If misdirected, please notify us by telephone and confirm
> that it has been deleted from your system and any copies destroyed.  If you
> are not the intended recipient you are strictly prohibited from using,
> printing, copying, distributing or disseminating this e-mail or any
> information contained in it.  We use reasonable endeavours to virus scan all
> e-mails leaving the Company but no warranty is given that this e-mail and any
> attachments are virus free.  You should undertake your own virus checking.
> The right to monitor e-mail communications through our network is reserved by
> us.
> _______________________________________________
> sunmanagers mailing list
> sunmanagers at sunmanagers.org
> http://www.sunmanagers.org/mailman/listinfo/sunmanagers
-- 
perl -e "print qq^bIG VeRN ! ^^qq^#'#Yv#=<D+ ^"
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From mot1 at btinternet.com  Wed Jan 19 12:17:43 2005
From: mot1 at btinternet.com (MOT ABU)
Date: Wed, 19 Jan 2005 17:17:43 +0000 (GMT)
Subject: Summary :How can you display the metrics in a Solaris routing table?
Message-ID: <20050119171743.74958.qmail@web86204.mail.ukl.yahoo.com>

Hello
 
Looks like I cannot view the WinXP pro Metric equivalent on Solaris.
Metric being the cost of the network, the higher the number the higher the cost.
 
Some of you were annoyed by my question. 
Sorry if I upset anyone.
 
Thanks to everyone for their suggestions.
 
 
Thanks to all including
Sean Timmins
Hutin Bertrand
--Drew.
Angelo
 
 
__________________________________________________________
How can you display the metrics in a Solaris routing table?

For example in WinXp I can execute 
route print
Which will display the metric in the right most column.

Thanks. 


		
---------------------------------
 ALL-NEW Yahoo! Messenger - all new features - even more fun!  
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From warren.liang at cox.net  Wed Jan 19 12:57:23 2005
From: warren.liang at cox.net (Warren Liang)
Date: Wed, 19 Jan 2005 12:57:23 -0500
Subject: SUMMARY: Solaris 10 and Network Appliance F840 filer
Message-ID: <20050119175723.TPRQ21241.fed1rmmtao09.cox.net@smtp.west.cox.net>

Good Morning,

Thanks to Joe Astrologo. I modified /etc/default/nfs, changed parameter NFS_CLIENT_VERSMAX from 4 to 3. It solves the problem for now.
I am working on downloading ONTAP 6.5.1R2P14.

Enclosed please review Joes reply:

Solaris 10 comes with native NFS client version 4. There is a known bug with Netapp Filers .
The filers can't handle the psuedo NFS mounts features of version 4 and panic..
Workaround is to stipulate nfs version 3 in /etc/default/nfs file on Solaris 10 box.
Tou can also upgrade your version of ONTAP to 6.5.1R2p14 i believe, which has the fix.
Joe Astrologo

My original posting is:

Good Morning,
I have a Solaris 10 on an Ultra 5 box. Try to mount a share on Netapp F840 filer, it crashes the filer.
Sun O/S:
SunOS solaris06 5.10 s10_72 sun4u sparc SUNW,Ultra-5_10
Filer:
NetApp F840
NetApp Release 6.4.2: Mon Sep 15 13:48:09 PDT 2003
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From akanellopoulos at upccorp.com  Wed Jan 19 15:53:24 2005
From: akanellopoulos at upccorp.com (Kanellopoulos, Angelos)
Date: Wed, 19 Jan 2005 21:53:24 +0100
Subject: SUMMARY urgent tar help
Message-ID: <E892FA932488464C8A483EB0BAF508C401FD5BAA@nlsrbacoexc06>

What to say about you great inteligent people, thank you so much for your
fast help-responses...Many many people replied , most of the replies were
refering using ln which works in many cases, not on mine.. i will post the
solution that worked for me and I think the most complete.
Respect to you my friends. Thank you again for your interest.

That Used to be my problem : 

-Hello again Guru's.. 
-I have a tar issue, please help me if you can. This is the issue : 
-I have a big tar file that has been tared with the -c (not -C) option,
--with the following path e.g. /home/user1 . The problem is the /home
-filesystem is full I need to extract it under e.g /usr/u01 (different fs) i
mean change the relative path that has been tared, without extrating -on
-another system and them movin. 


>   There are a couple of alternatives: 
> 
> a. Method 1 (user) 
>                 1. /usr/bin/pax -r -s ',^/,,' -f file.tar 
b. Method 2 
> (root) 
>                 1. /usr/bin/cp /usr/sbin/static/tar /tmp 
>                 2. /usr/bin/dd if=file.tar | /usr/bin/chroot /tmp 
> ./tar xf - 
> 
> or.... 
> 
> mkdir /var/tmp/new/ 
> cp /usr/sbin/static/tar /var/tmp/new 
> cp file.tar /var/tmp/new 
> chroot /var/tmp/new ./tar new.tar 

Frank Huang 

Other many other people replied the same as frank i just posted his which
includes the rest plus the ln,
so pex , chroot, ln is the way to do it.
for the last time Thank you people.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From solaris_admin-1 at yahoo.com  Thu Jan 20 00:19:53 2005
From: solaris_admin-1 at yahoo.com (solaris_admin-1 at yahoo.com)
Date: Wed, 19 Jan 2005 21:19:53 -0800 (PST)
Subject: SUMMARY: Problems with Automount
In-Reply-To: <20050120043731.58717.qmail@web14021.mail.yahoo.com>
Message-ID: <20050120051954.62235.qmail@web14023.mail.yahoo.com>

Thanks a lot to Kelly Ormsby's help..also thanks to
all the "out of office emails" =)

I figured out the problem after running automount in
debug mode....

The problem was due to the use of virtual hostnames
(multiple hostnames/IP addresses). TCP had a problem
binding to the correct IP address. 

Thanks Again....

--- solaris_admin-1 at yahoo.com wrote:

> Hi --
> 
> i am having problems with Automount on one of my
> Solaris Servers.
> 
> /etc/init.d/auto_home is configured, 
> /etc/init.d/rpc is started
> /etc/init.d/nfs.client is started
> /etc/init.d/nfs.server is started
> /etc/init.d/autofs is started
> 
> I cannot even automount a local home directory
> (example: user localhost:/export/home/user). It just
> hangs. Nothing in /var/adm/messages..
> 
> Any Ideas? 
> 
> Thanks
> _______________________________________________
> 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 rachid.boukhari at st.com  Thu Jan 20 05:46:13 2005
From: rachid.boukhari at st.com (Rachid BOUKHARI)
Date: Thu, 20 Jan 2005 10:46:13 +0000
Subject: Summary: Different mount options in NIS map
In-Reply-To: <41E69419.5060605@st.com>
References: <41E69419.5060605@st.com>
Message-ID: <41EF8BF5.90107@st.com>

Hi,

Sorry for the delay;
well I have no other solution except to create special netgroups for 
this purpose .

rachid boukhari wrote:

> Hi Gurus,
>
> I would like to makea  NIS map "ro" mounted for some clients and "rw" 
> for others. How can I procede.?
>
> rdc_23     ro,nosuid,intr,hard     neptune:/soft/rdc/23
>
> What should change if possible in this syntax.
>
> I'll summarize .
>
> Rachid,
>

-- 
****************************************
* Rachid BOUKHARI                      *
* Unix/Linux System Administrator      *
* Rabat Information Security Officer   *
* STMicroelectronics-Rabat             *
*                                      *
* BP 765 Av IbnSina Agdal Rabat-Morocco*
* Phone: +212 22 93 63 29              *
* Fax: +212 22 93 63 34                *
****************************************
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From rossman at columbia.edu  Thu Jan 20 09:05:18 2005
From: rossman at columbia.edu (Ken Rossman)
Date: Thu, 20 Jan 2005 09:05:18 -0500
Subject: SUMMARY: Locating process that opens a TCP or UDP listener
In-Reply-To: <453D335C35136F4B8163C281D58BBD884D10A8@NBNUNYCEXCH1.nesbittburns.ca>
References: <453D335C35136F4B8163C281D58BBD884D10A8@NBNUNYCEXCH1.nesbittburns.ca>
Message-ID: <5096C179-6AEC-11D9-89F2-000D9331DA10@columbia.edu>

Thank you all for the immediate and volumunous responses I got to this
my question about locating the PID owning a given network port.

The near-unanimous response was to use 'lsof', or more specifically:

   # lsof -i TCP:<port#>

(or some variation of the above)

Now I just need to find a way to compile a 32-bit SPARC version of lsof
to run on an early Ultra-1 that refuses to run the 64-bit kernel... :-)

Thanks to all,
Ken Rossman
rossman at columbia.edu
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From warren.liang at cox.net  Thu Jan 20 11:22:21 2005
From: warren.liang at cox.net (Warren Liang)
Date: Thu, 20 Jan 2005 11:22:21 -0500
Subject: SUMMARY: sendmail fails after applied Solaris 8 Recommended Patch
Message-ID: <20050120162223.IAIA27771.fed1rmmtao08.cox.net@smtp.west.cox.net>

Good Morning,

Thanks for many valuable responses. I backout the patch and sendmail works finn now.

I like to share a couple of great solution/recommendations from Allen and Erek:

Allen wrote:
Looks like the patch cluster replaced your custom sendmail binary.. 
try doing /usr/lib/sendmail -bt -d0.1 and look for MILTER support. 

You may have to rebuild sendmail from source with MILTER support.
If you do.. i'd remove the SUNWsndmr and SUNWsndmu packages so
future patches won't replace the binary.

Erek wrote:

Sun makes a couple of assumptions about patches:

* You have never changed your system from the "Out of the Box" defaults
* You would never install a third party application

IOW, they simply dropped a new sendmail binary on top of your Miltercompiled one.  You'll have to rebuild sendmail + milter + mimedefangbinary again.  And as a hint--Never place your sendmail binary over Sun's.
:)  Always install into /usr/local or /opt or /whatever/your/heart/desires
.  Just not in /usr/lib/ !

If you update and change from the standard Sun sendmail, you might want to
consider removing the sendmail packages from your server.  That way
Solaris won't even try to patch sendmail since it doesn't exist on your
server.

My original posting:

sendmail fails after applying Solaris 8 Recommended patch.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From guenthere at yahoo.com  Thu Jan 20 11:59:10 2005
From: guenthere at yahoo.com (Ed Guenther)
Date: Thu, 20 Jan 2005 08:59:10 -0800 (PST)
Subject: SUMMARY: moving a boot drive 
Message-ID: <20050120165910.70801.qmail@web40905.mail.yahoo.com>

If it can be done no one here knows how.  I've tried
various ideas, nothing worked.  Once an x86 system is
installed the boot disk is stuck to that drive bay.




		
__________________________________ 
Do you Yahoo!? 
Read only the mail you want - Yahoo! Mail SpamGuard. 
http://promotions.yahoo.com/new_mail 
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From jastrologo at factset.com  Wed Jan 19 13:27:10 2005
From: jastrologo at factset.com (jastrologo at factset.com)
Date: Wed, 19 Jan 2005 13:27:10 -0500
Subject: Instant Messaging client for Solaris sparc 8/9 (Summary)
In-Reply-To: <OF3779E745.472A55DC-ON85256F8E.00620B1B-85256F8E.00624C02@factset.com>
Message-ID: <OFAC66EB9D.F6AA4E0B-ON85256F8E.00650700-85256F8E.00655D44@factset.com>

In summary-- Its sounds like the majority consensus  is to use gaim.

Thanks for all who replied.


Joe Astrologo


Unix Systems Engineer


Factset Research Systems


Voice: 203-863-1562


                                                                           
             jastrologo at factse                                             
             t.com                                                         
             Sent by:                                                   To 
             sunmanagers-bounc         sunmanagers at sunmanagers.org         
             es at sunmanagers.or                                          cc 
             g                                                             
                                                                   Subject 
                                       Instant Messaging client for        
             01/19/2005 12:53          Solaris sparc 8/9                   
             PM                                                            
                                                                           
                                                                           
                                                                           
                                                                           
                                                                           




Hello--

Are there any recommended instant messaging clients available for the
Solaris 8/9 sparc platform?

Sun Java Instant messaging seems to only have free downloads for linux and
windows paltforms.

Thanks in advance.



Joe Astrologo


Unix Systems Engineer


Factset Research Systems


Voice: 203-863-1562
_______________________________________________
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 suresh at ind.tansu.com.au  Fri Jan 21 00:24:15 2005
From: suresh at ind.tansu.com.au (Suresh Bangalore)
Date: Fri, 21 Jan 2005 16:24:15 +1100 (EST)
Subject: SUMMARY: SunFire - V250 Boot problem 
In-Reply-To: <Pine.GSO.4.44.0501201708390.25791-100000@laptop-iserver-1.ind.tansu.com.au>
Message-ID: <Pine.GSO.4.44.0501211619480.25791-100000@laptop-iserver-1.ind.tansu.com.au>

Hi All

Thanks for all for the replies.

> ERROR: The following devices are disabled:
>     scsi
>     net

Thanks to Michael for suggesting to use asr-clear
At the OK prompt,
asr-clear clears all the disabled devices

> INIT: Cannot create /var/adm/utmpx
>
>
>                               INIT: SINGLE USER MODE
>
>                                                     Type control-d to
> proceed with normal startup,
>                   (or give root password for system maintenance):

Surprisingly one of our developers had completly screwed up /etc/inittab
file. Its normal to look at things like file system full, permissions,
ownerships etc, but had never thought about /etc/inittab. Well, learning
new things everyday.

Thanks for all the help.
Cheers
Suresh Bangalore
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From wolf at uen.org  Fri Jan 21 11:04:33 2005
From: wolf at uen.org (Wolfgang Schwurack)
Date: Fri, 21 Jan 2005 09:04:33 -0700
Subject: SUMMARY: Disaster Recover
Message-ID: <41F12811.7020009@uen.org>

I had a lot of good replies. The one that I liked the best was from Ed 
Alexander who sent me a script that gave me all of the information I 
needed to recover from a disaster. see attachment

Some other options

from przemol

Look at the explorer:
http://sunsolve.sun.com/pub-cgi/show.pl?target=explorer/explorer

from Aleksander Pavic
http://www.sun.com/bigadmin/scripts/indexBackup.html

from Prasad Dharmasena

As root, run these to get quite a lot of info about a machine.

/usr/bin/hostname
/usr/bin/domainname
/usr/bin/uname -a
/usr/platform/sun4u/sbin/prtdiag -v
/usr/bin/getconf -a
/etc/mount
/usr/bin/df -k
/usr/bin/netstat -i

>From Bob Bumpus

I've used cfg2html to collect system information.  You may want to look 
at it.

http://come.to/cfg2html



To All

I am preparing for a Disaster Recovery plan for all of my solaris 
server. One of the first steps is to create a script that will collect 
the storage structure and other specification of the system.  Does 
anyone have a script or know of a program the will get all of the 
information that one would need if a server fails?

thanks
-- 

  /\         /\     \   O \    Wolfgang Schwurack
 / \\  /\   /\\\     \_/|\_\   Unix System Administrator
/   \\/ \\ /  \\\       \\     University of Utah/Utah Education Network
     /   \/    \\\     \/'     Tel: (801) 587-9444
    /    /      \\\     \\_    email: wolf at uen.org
   /    /        \\\     -
#!/bin/sh
#==============================================================================
# $Id: sysconfig,v 1.1 1998/12/01 08:42:23 peters Exp $
#
# NAME: 	sysconfig
# DESCRIPTION:	Shows a summary of useful configuration information on 
#		Solaris systems.
# DATE:		2 Dec 1998
# AUTHOR:	Peter Sundstrom (peters at ginini.com.au)
# SOURCE:	http://www.ginini.com.au/tools/sysconfig
#
#==============================================================================

USAGE="Usage: `basename $0` [-Dcdgkhmnprst]"

SCRIPT=`basename $0` # Shortname of this script
RELEASE=`uname -r` # OS Release
PLATFORM=`uname -s` # Machine platform

#------------------------------------------------------------------------------
help ()
{
  cat <<EOF

$USAGE

Displays useful configuration information on a Solaris system.

-D Debug mode
-c Crontab files
-d Available Disks
-g General machine configuration
-h Display help
-k Installed packages
-m Mounts and disk usage
-n Network interfaces
-p Disk partitions
-r RAID Metadata
-s Swap space
-t Installed patches

If no options are specified, all information is displayed.
EOF

  exit
}

#------------------------------------------------------------------------------
General_info ()
{
  [ "$DEBUG" = TRUE ] && set -x

  echo "==============================================================================="
  echo "==========General Information=================================================="
  echo "===============================================================================\n"

  showrev | egrep -v "Release"
  release=`grep Solaris /etc/release`
  echo Release: $release
  echo "Machine type: \c"
  uname -i

  prtconf | grep "Memory size"

  maxprocessors=`psrinfo | wc -l`
  echo Number of processors: $maxprocessors

  if [ $maxprocessors -gt 1 ]
  then
     processor=0
     while [ $processor -lt $maxprocessors ]
     do
       speed=`psrinfo -v $processor| grep "operates at" | sed "s/.*operates at //"`
       echo "Processor $processor: $speed"
       processor=`expr $processor + 1`
     done
  else
       speed=`psrinfo -v | grep "operates at" | sed "s/.*operates at //"`
       echo "Processor runs at $speed"
  fi

  return 0
}

#------------------------------------------------------------------------------
Crontab_Files ()
{   
  [ "$DEBUG" = TRUE ] && set -x

  echo "\n"
  echo "==============================================================================="
  echo "==========Crontabs============================================================="
  echo "===============================================================================\n"

  for ctab in /var/spool/cron/crontabs/*
  do
    echo "===Contents of ${ctab}:\n"
    /usr/bin/crontab -l `/usr/bin/basename ${ctab}`
    echo "\n" 
  done

  return 0
}

#------------------------------------------------------------------------------
Swap_space ()
{   
  [ "$DEBUG" = TRUE ] && set -x

  echo "\n"
  echo "==============================================================================="
  echo "==========Swap Space==========================================================="
  echo "===============================================================================\n"
  echo "Swap device               Size"
  echo "-------------------------------"  
  swap -l | grep -v swapfile |
  while read swapfile dev swaplo blocks free
  do
    size=`expr $blocks / 2048`
    echo "$swapfile\t$size Mb"
  done

  return 0
}

#------------------------------------------------------------------------------
Mounts_and_Drive_Space ()
{
  [ "$DEBUG" = TRUE ] && set -x

  echo "\n"
  echo "==============================================================================="
  echo "==========df and vfstab========================================================"
  echo "===============================================================================\n"
  echo ""

  echo "===df -kl Output:"
  /usr/bin/df -kl

  echo "\n"

  echo "===Contents of /etc/vfstab:\n"
  /usr/bin/cat /etc/vfstab

  return 0
}

#------------------------------------------------------------------------------
Network_interfaces ()
{
  [ "$DEBUG" = TRUE ] && set -x

  echo "\n"
  echo "==============================================================================="
  echo "==========Network Interfaces==================================================="
  echo "===============================================================================\n"
  echo "Int      Host     Address"
  echo "-------------------------------------------------------------------------------"

#  typeset -L7 interface
#  typeset -L8 intname

  for interface in `ls /etc/hostname.*|sed "s/.*\.//"`
  do
    intname=`cat /etc/hostname.$interface`
    echo "$interface  $intname \c"
    ifconfig $interface | grep inet | sed "s/^.*inet //"
    ifconfig $interface | grep ether
  done

  return 0
}

#------------------------------------------------------------------------------
Disks ()
{
  [ "$DEBUG" = TRUE ] && set -x

  echo "\n"
  echo "==============================================================================="
  echo "==========Available Disks======================================================"
  echo "===============================================================================\n"
  disks=`echo "0\nq"|format 2>/dev/null |grep "[0-9]\."|sed "s/^ *//"`
  echo "$disks"

  return 0
}


#------------------------------------------------------------------------------
Disk_partitions ()
{
  [ "$DEBUG" = TRUE ] && set -x

  echo "\n"
  echo "==============================================================================="
  echo "==========Disk Partitions======================================================"
  echo "===============================================================================\n"

  disks=`echo "0\nq"|format 2>/dev/null |grep "[0-9]\."|sed "s/^ *//"`
  for disk in `echo "$disks" | awk '{print $2}'`
  do
    echo "\nDisk $disk"
    echo "--------------"
    prtvtoc -s /dev/rdsk/${disk}s2
  done

  return 0
}

#------------------------------------------------------------------------------
RAID_Metadata ()
{
  [ "$DEBUG" = TRUE ] && set -x

  if [ -d /etc/lvm ]
  then
    echo "\n"
    echo "==============================================================================="
    echo "==========RAID Metadata========================================================"
    echo "===============================================================================\n"
    echo ""

    for mfile in md.cf md.ctlrmap md.tab mddb.cf
    do
      if [ -f /etc/lvm/${mfile} ] 
      then
        echo "\n===Contents of /etc/lvm/${mfile}:\n"
        cat /etc/lvm/${mfile}
        echo "\n"
      fi
    done
  fi

  return 0
}

#------------------------------------------------------------------------------
Installed_packages ()
{
  [ "$DEBUG" = TRUE ] && set -x

  echo "\n"
  echo "==============================================================================="
  echo "==========Installed Packages==================================================="
  echo "===============================================================================\n"
  
  pkginfo

  return 0
}

#------------------------------------------------------------------------------
Installed_patches ()
{
  [ "$DEBUG" = TRUE ] && set -x

  echo "\n"
  echo "================================================================================"
  echo "==========Installed Patches====================================================="
  echo "================================================================================\n"

  showrev -p

  return 0
}
  
#------------------------------------------------------------------------------
# END OF FUNCTIONS
#==============================================================================

#
# Some functions need root access
#
if [ `id|sed -e "s/uid=//" -e "s/(.*//"` != 0 ]
then
   echo "$SCRIPT needs to be run as root" >&2
   exit 1
fi

#
# Check to see what platform we are running on.
#
if [ $PLATFORM != SunOS ]
then
   echo "$SCRIPT can only run on Sun systems" >&2
   exit 1
fi

#    
# Check to see if we are on Solaris 2.5 (SunOS 5.5) or higher
#
if echo $RELEASE | grep 5.[5-9] >/dev/null 2>&1
then
   : # OS level is OK
else
   echo "$SCRIPT needs to be run on Solaris version 2.5 and higher" >&2
   exit 1
fi

#
# Set an appropriate path
#
PATH=/usr/bin:/usr/sbin 

#
# If no parameters are supplied, set flag to indicate to display 
# all information.
#
[ $# -eq 0 ] && ALL=TRUE

echo "System configuration as of `/usr/bin/date`"
echo ""
echo "Generated by /home/systems/scripts/${SCRIPT} $*"
echo ""

while getopts :Dcdghkmnprst opt
do 
  case $opt in
  D) DEBUG=TRUE
set -x
;;
  c) Crontab_Files
;;
  d) Disks
;;
  h) help
;;
  g) General_info
;;
  k) Installed_packages
;;
  m) Mounts_and_Drive_Space
;;
  n) Network_interfaces
;;
  p) Disk_partitions
;;
  r) RAID_Metadata
;;
  s) Swap_space
;;
  t) Installed_patches
;;
  ?) echo "$USAGE" >&2
exit 1
;;
  esac
done

#
# Exit if flags have been specified
#
[ "$ALL" = TRUE ] || exit 0

#
# By default, display all information
#
General_info
Crontab_Files
Swap_space
Mounts_and_Drive_Space
Network_interfaces
Disks
Disk_partitions
RAID_Metadata
Installed_packages
Installed_patches
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From steve.rieger at tbwachiat.com  Mon Jan 24 13:20:06 2005
From: steve.rieger at tbwachiat.com (steve rieger)
Date: Mon, 24 Jan 2005 13:20:06 -0500
Subject: SUMMARY need advice and feedback
Message-ID: <41F53C56.308@tbwachiat.com>

the answers were split up as follows.

about 45% of production systems, had the sw compiled on a dev box from 
source and then moved it to prod.


the other 45% used sunfreeware, or blastwave for all if not most of 
their s/w

two responded stating that they use other sw pkg's,

other responses were, openpkg, only sunw packages, and pkgsrc.



I was also asked if this will be made available online, and the answer 
is yes. i will not post your names or direct responses online. as soon 
as i have it online i will let you know. i just wanted to get a feel for 
how other admins do their thing. please do send a response if you can, 
if you did not do so.



thank you all for your feedback






with the question being


With this being a list of admin who, after being a member for more than
4 years, i grew to respect your knowledge, i would like to know how you
prefer to install software on solaris. is it from sunfreeware, netbsd's
pkgsrc, openpkg, or from source code.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From ahoesch at smartsoft.de  Mon Jan 24 17:17:37 2005
From: ahoesch at smartsoft.de (=?ISO-8859-1?Q?Andreas_H=F6schler?=)
Date: Mon, 24 Jan 2005 23:17:37 +0100
Subject: Summary: postfix mail aliases
Message-ID: <C14EF5C6-6E55-11D9-A61F-000393CA0072@smartsoft.de>

Dear managers,

thanks to

1) "Webmaster - ITServ GmbH" <webmaster at itserv.de>
2) Marian Dragomir <marian at romsys.ro>
3) "Michael Schneider/calispera.com" <michael.schneider at calispera.com>
(and many other, I accidentally sent the summary to Michael instead of 
the list)

for their kind feedback.

1) suggested to look for a .forward file in the home dir. There was 
none.
2) suggested to run "postmap /etc/mail/aliases" but this gave me

postmap: warning: /etc/mail/aliases, line 1: record is in "key: value" 
format; is this an alias file?
...

Not sure what this command expects.

3) finally suggested to rebuild the NIS maps. This actually solved the 
problem.

Thanks a lot!!!

Regards,

   Andreas

> I just removed an alias
>
> dherbst: pkopp
>
> from /etc/mail/aliases and run
>
> newaliases
>
> I also did "postfix reload" and "postfix stop", "postfix start", but
> mails to dherbst are still forwarded to pkopp.
>
> Jan 24 12:27:54 centrino postfix/local[23699]: [ID 197553 mail.info]
> 99A1F1034: to=<pkopp at bockelmann-holz.com>,
> orig_to=<dherbst at bockelmann-holz.de>, relay=local, delay=0, status=sent
> (mailbox)
>
> What am I missing here?
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From rob.de.langhe at belgacom.be  Tue Jan 25 03:29:10 2005
From: rob.de.langhe at belgacom.be (rob.de.langhe at belgacom.be)
Date: Tue, 25 Jan 2005 09:29:10 +0100
Subject: SUMMARY: adding EMC-CX600 LUN: format gives "drive type
  unknown"
Message-ID: <20E53B61AAB3994EA15681023EE52418DA0873@AE0007.BGC.NET>

Thanks all for the answers and suggestions, some of which confirmed what
I found out myself in the mean time: the array-support for Clariion was
not included by default in the VxVM installation ("vxddladm listsupport
|grep DGC")

So I downloaded from the Veritas web site the package "CLR-ASL", which
arranged that the disk seen through 2 HBAs, was properly identified by
VxVM as the single and same disk.


Some people remarked that we use Qlogic-2300 cards and suggested that
could have been the cause of the 'problem': let me confirm that I have
connected a large combination of SUN servers with different HBA types,
using different drivers, to EMC disk arrays, directly attached or via
SAN switches. It is only a matter of -indeed- properly veryfing if a
certain combination server/HBA/driver (with or without VxVM and/or EMC's
PowerPath) has been validated by EMC, and testing out the exact
installation order of packages, configuration files etc to get the thing
work. Often EMC could not indicate the precise order of installations or
configuration details themselves. It's just a bit of initial work to
find the right combinations, but then you're done for any further
machine that you install with the exact same combination.
Do never forget to do the ultimate test : force I/O to the LUNs (like a
"verify" in "format"), verify via which HBA the traffic is going, and
pull the plug from that HBA : the I/O should just block for a few
seconds, then continue over the other HBA. Powerpath is handy to know
what HBA-path is being used, but only the full-featured $$-version
provides dual-HBA load balancing of the I/O traffic, the normal
$-version gives just path-failover, which can only be done using
VxVM-DMP. Or the Solaris-bundled "mpxio", for that matter.

Cheers

Rob


-----Original Message-----
From: sunmanagers-bounces at sunmanagers.org
[mailto:sunmanagers-bounces at sunmanagers.org] On Behalf Of
rob.de.langhe at belgacom.be
Sent: 24 January 2005 13:57
To: sunmanagers at sunmanagers.org
Subject: adding EMC-CX600 LUN: format gives "drive type unknown"

Hi,

we have a E15K domain running Solaris-8, and connected to a SAN via
QLogic-2300 Host Bus Adapters.

A bunch of EMC-Symmetrix disks are already made available to the server
and being used for filesystems through VxVM.

Now we need to add an extra LUN coming from an EMC-Clariion CX600 array.
The new LUN is discovered with SCSI-probing, but showing up in "format"
as a "drive type unknown" ...

When I try to let "format" "auto-configure" the disk type (option
"type", then select "0. Auto-Configure") I get the message

Auto configure failed

Within "format", the "inquiry" option yields

Vendor: DGC
Product: CX700
Revision: HP03


In VxVM, I have added the JBOD id with

vxddladm addjbod vid=DGC pagecode=0x83 offset=8 length=16

The people here told me that in previous attempts to make new
Clariion-LUNs correctly identified by Solaris on a machine that has
already Symmetrix-LUNs, was only after a reboot ... Which is surely not
the procedure I want to adopt (production servers, hey)

Any suggestions what can be done about it ?

TIA

Rob


**** DISCLAIMER ****
http://www.belgacom.be/maildisclaimer
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


**** DISCLAIMER ****
http://www.belgacom.be/maildisclaimer
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From gosselinm at netscout.com  Tue Jan 25 11:19:52 2005
From: gosselinm at netscout.com (Gosselin, Mark)
Date: Tue, 25 Jan 2005 11:19:52 -0500
Subject: Summary: DiskSuite question
Message-ID: <9E164A8DC7C9464688E04C3DBE2E56585092B2@nsxmail.netscout.com>

Hi All,
 
Thanks to Daniel Nuno for the only response. He pointed me to http://www.unixway.com/vm/solarisvm.html which provided me with what I needed.
Simple process:
 
Replace hard drive, reinstall OS, install DiskSuite, restore /etc/lvm/* from tape, restore lines to /etc/vfstab that mount metadevices... Reboot. That was
it. Some mirrored metadevices took a while to initialize because of the way they went down, but other than that, it was very simple...
 
Thanks again,
 
Mark Gosselin
NetScout Systems

-----Original Message-----
From: Gosselin, Mark 
Sent: Sunday, January 23, 2005 5:11 PM
To: sunmanagers at sunmanagers.org
Subject: DiskSuite question


Hi All,
 
I had a boot disk failaure and want to restore my attached disk arrays (configured in DiskSuite). I have good backups of all files from the system. Is there a document or procedure that will make retoring the DiskSuite config easy???
 
Thanks,
Mark Gosselin
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From sreeredhat at yahoo.com  Tue Jan 25 14:30:34 2005
From: sreeredhat at yahoo.com (sreenath sarikonda)
Date: Tue, 25 Jan 2005 11:30:34 -0800 (PST)
Subject: Summary: netstat -rn output changing automatically
Message-ID: <20050125193034.31771.qmail@web54207.mail.yahoo.com>

>Hell Gurus,
>           I was checking the routing table of my
>system and it was chaning automatically. I have
>/etc/notrouter. Could some one explain this to me? 

First of all thank you to Russell Page, Bruce
Kirkland, Ryan Krenzischek, Crist Clark,
Matthew Stier and jzhao.
My problem was solved by removing the
/etc/resolv.conf. I really don't know whether my
following theory is correct. The name server it was
pointing doesn't have any information regarding our
local systems and its not a local DNS server. I guess
the system was wasting time in DNS queries. We don't
have local dns server.As soon as I removed the file,
ftp connections stopped dropping.
Couple of points:
/etc/notrouter only disables ipforwarding.
It will still startup startup router discovery, if
/etc/defaultrouter 
is not set.
___________________________________
About parameter :ip_ignore_direct is set to zero
Then if you want to prevent those routes from coming
back, you need to set it to one. The "D" in the
"Flags"
column indicates these routes are from redirects.
Redirects are supposed to improve performance. You may
want to look into why they are being sent in the first
place. This may be an indication that your network is
misconfigured.

Very much appreciated Russell's summary. 
There are three different concepts to get in order to
understand 
routing.

1. Is this machine a router?
2. Where should this machine forward packets to?
3. How does this machine acquire the routing
information it needs in 
order 
to answer question 2?

1. An IP router is ANY system that is connected to
more than one IP 
network, 
that forwards traffic between the networks. A Solaris
system with more 
than 
one configured interface will automatically configure
itself as a 
router. 
This means that other machines can forward IP packets
to it to be 
forwarded 
to another network. The file you mention -
/etc/notrouter - suppresses 
this 
behaviour. For instance a busy database server may be
connected to two 
or 
three networks. If it starts acting as a router then
it will be 
sacrificing 
resources to the routing function. In fact if there is
a lot of traffic 
to 
route, it may end up spending more time routing IP
traffic than acting 
as a 
database server! By creating /etc/notrouter, we
prevent it from 
configuring 
itself as a router.

2. Where should this machine forward packets to?
All systems connected to a network can deliver IP
traffic directly to 
other 
systems on the same subnet. If the destination IP
address is on a 
remote 
network, the system will deliver the traffic to a
router. The system 
consults the routing table to determine which router
it should forward 
a 
particular IP datagram to. Routing tables typically
contain one or more 
of 
three types of entries.
A network route specifies which router to choose to
forward traffic 
directed 
to a particular IP network. A host route specifies a
router for traffic 
going to a particular host, and the (usual) default
route specifies a 
router 
for all traffic that is not specified by a net or host
route.

3. How does this machine acquire the routing
information it needs in 
order 
to answer question 2?
There are three sources a machine can use to acquire
routing 
information. 
The information can be entered into the routing table
by a system 
administrator or read from a file. A very common case
is of a system 
that is 
connected to a subnetwork with only one router on it.
Obviously all 
remote 
traffic must be sent to this router. If we create a
file called 
/etc/defaultrouter containing the local IP address of
the router, this 
file is read by the system at boot time, a default
route is added to the 
routing table, and it remains there until it is
manually removed or the system 
shuts down. No further, automatic updates to the
routing table occur.
A second way to learn where the default router is, is
for the system to 
broadcast a special request called a router
solicitation message when 
it boots. Many routers, and Solaris systems running as
routers respond to 
these messages. If the system receives a response it
updates it's routing 
table with a default route. This will remain in the
table as long as the 
router continues to provide the service. If the
default router "disappears" 
the host will broadcast a new solicitation message to
find a new router.
The third way for the system to learn about routers is
to monitor 
"Router Information Protocol" broadcasts. Essentially,
every 30 seconds or so 
routers dump their routing tables into a UDP packet
and broadcast it to 
all attached networks. Hosts that are listening for
these RIP broadcasts 
update their routing tables automatically as they
learn new information about 
the network.
My experience is that most Solaris administrators
either create an 
/etc/defaultrouter file or depend on the router
discovery protocol to 
configure a default route for them.
Your post suggests that your system, if it has more
than one interface, 
is not itself acting as a router  - you have an
/etc/notrouter file, which 
presumably was there last time the system booted - and
that it is 
listening to, and acting on RIP broadcasts to maintain
it's routing table.

Thank you,
Sree.


		
__________________________________ 
Do you Yahoo!? 
Yahoo! Mail - Find what you need with new enhanced search.
http://info.mail.yahoo.com/mail_250
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From Scott.Cooper at edl.uk.eds.com  Wed Jan 26 09:52:28 2005
From: Scott.Cooper at edl.uk.eds.com (Scott.Cooper at edl.uk.eds.com)
Date: Wed, 26 Jan 2005 14:52:28 -0000
Subject: SUMMARY: UFS Logging - disadvantages?
Message-ID: <6C5F0F3099DC5B45B87D78EE2A37830D0F3245@ukhkm001.edsadewd.edl.uk.eds.com>

Thanks to all those that responded!
 
So many replies, seems the advantages far outweigh any performance deficit
especially in our environment.
 
I included this reply as it was by far my favourite ( Credit to Michael
Horton!)
 
====================
 
no calls to fsck disks during reboots will make you look unimportant.

no rescue beeps in the night that get you out of awkward situations.

no ignoring the boss with "i've got to fix the server!".

 

but this can be remedied with cron and other programs.

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

Cheers all!

Scott

Original message:

Hi all,

Looking at implementing UFS Logging on our servers by adding the logging
entry in /etc/vfstab, the servers can have a habit of unscheduled poweroffs
and with potentially huge chunks of data attached logging enabled could give
huge benefits in revovering a system like this.

Are there any disadvantages to using it?
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From sreeredhat at yahoo.com  Wed Jan 26 14:43:52 2005
From: sreeredhat at yahoo.com (sreenath sarikonda)
Date: Wed, 26 Jan 2005 11:43:52 -0800 (PST)
Subject: SUMMARY : ftp permission denied to write into folders created
 by the ft p session.
Message-ID: <20050126194352.38155.qmail@web54207.mail.yahoo.com>

/apps was the mount point. I changed its permissions
and remounted and it worked. If you try to delete a
directory without aforementioned changes then you get
the following error
rm: cannot determine if this is an ancestor of the
current working directory

Thank you,
Sreenath.

> Hello Guru's,
>               Follwing the solaris hardening
> priciples
> I set the default umask as 023 for the files in
> /etc/default/login file for our new systems. I am
> able
> to  create directories using ftp session. But I am
> not
> able to further descend into the diretories created
> by
> ftp session, in ftp session. However I am able to
> descend into these directories using login shell.
> Here
> are the directory permissions.
> ls -ld /apps/commonfund/  (user home direcotry) 
> drwxr-xr--   9 user1 somegroup    /apps/commonfund//
> 
> ls -l /apps/commonfund/ 
> -rw-r--r--   1 user1 somegroup     local.profile
> drwxr-xr-x   3 user1 somegroup     pace/
> drwxr-xr--   2 user1 somegroup     pace2/
> drwxr-xr--   2 user1 somegroup     pace3/
> 
> I created pace,pace2 directories withing ftp
> session.
> But I am not able to write/read into them. Any help
> would be appreciated. I trimmed the output.
> 
> Thank you,


		
__________________________________ 
Do you Yahoo!? 
The all-new My Yahoo! - Get yours free! 
http://my.yahoo.com 
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From StephenB at datacom.co.nz  Wed Jan 26 16:58:15 2005
From: StephenB at datacom.co.nz (Stephen Barnett (DSL AK))
Date: Thu, 27 Jan 2005 10:58:15 +1300
Subject: Summary: SSH hangs from cron on Solaris 9
Message-ID: <A02D45F2829A894C9B186090DCF8D2FC24EFE7@dnzakex1.datacom.co.nz>

Hi,

It turns out that it's a stdin issue. The following command works:

echo "mkdir -p /mnt/backup/sxsppdb01" | ssh sxsprdb01

Thanks to Doc G, you weren't too far off the mark afterall.

Cheers,

Stephen

-----Original Message-----
From: sunmanagers-bounces at sunmanagers.org
[mailto:sunmanagers-bounces at sunmanagers.org] On Behalf Of Stephen
Barnett (DSL AK)
Sent: Thursday, 27 January 2005 10:13 a.m.
To: sunmanagers at sunmanagers.org
Subject: SSH hangs from cron on Solaris 9

Hi,

I have just discovered that a command that runs fine from command line:

ssh sxsprdb01 "mkdir -p /mnt/backup/sxsppdb01"

Hangs when run from cron.

SSH 3.9p1 / Solaris 9 Generic_117171-09

Any thoughts?

Thanks,

Stephen W. Barnett
Systems Consultant - Unix Division
Datacom Systems Limited
Auckland
New Zealand
_______________________________________________
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 Francois.Bousquet at cgi.com  Wed Jan 26 13:46:18 2005
From: Francois.Bousquet at cgi.com (Bousquet Francois)
Date: Wed, 26 Jan 2005 13:46:18 -0500
Subject: SUMMARY: 64 bits version of the find command
Message-ID: <1F85129C88B3714185C912F3AF47605012A2592C@nc1981exc03.banquelaurentienne.ca>

I finally solve my problem :

The problem is the /proc pseudo file system which may contain some 64 bits
process information.  These files can't be accessed by the default find
command available in Solaris 7.  This explain the following error message.
find: cannot open /: Value too large for defined data type

find misses a exclude options, so it is not possible to prevent it from
scanning the /proc file system.  The only things that may be done are to
grep the output or use the fstype ufs options, but these options will not
prevent find to scan /proc but only not return any of its content.   find
was then still scanning /proc and getting an error.

I then manage to get the source of GNU find @
http://www.gnu.org/software/findutils/findutils.html

and then compile find for sparc 64 bits with gcc 3.3    

The new find 64 bits command is now able to scan everything.

Some errors may output in /proc, but these error are only related to unlink
files.  So I am using 2>/dev/null after the find command to prevent this.



-----Original Message-----
From: Bousquet Francois [mailto:Francois.Bousquet at cgi.com]
Sent: January 21, 2005 11:11 AM
To: Mailing List - SunManagers (E-mail)
Subject: 64 bits version of the find command


I am trying to run a script that is doing a find for suid and guid files on
a system.

Basically I am using this command :

find / -perm 4000 -print


On one of my server (E450 with Solaris 7) this command returns an error :
# find / -perm 4000 -print
find: cannot open /: Value too large for defined data type
#

I have investigate the problem a bit more and found that it is /proc the
problem
# find /proc -perm 4000 -print
find: cannot open /proc: Value too large for defined data type
#

I also found this bug ID 4638523
http://sunsolve.sun.com/search/document.do?assetkey=1-1-4638523-1&searchclau
se=%22Value%20too%20large%20for%20defined%20data%20type%22%20%20%20find

The conclusion is proposing a script : find / ! \( -name proc -prune \) to
solve the problem.  But it is buggy, as it will exclude proc files or
directory on a deep directory tree (ex : /opt/application/proc).

So I am trying to find a 64 bits version of find for Solaris 7.
These are the only 64 bits command I can find :

# ls -l /usr/bin/sparcv9
total 1380
-r-xr-xr-x   1 bin      bin       192488 Jun 29  1999 adb
-r-xr-xr-x   1 bin      bin        19640 Sep  1  1998 gcore
-r-xr-sr-x   1 bin      sys        15480 Sep  1  1998 ipcs
-r-xr-xr-x   1 bin      bin        27296 Sep  1  1998 kgmon
-r-xr-xr-x   1 bin      bin        23184 Sep  1  1998 ls
-r-xr-xr-x   1 bin      bin        14584 Sep  1  1998 plimit
-r-xr-xr-x   1 bin      bin        81536 Sep  1  1998 prex
-r-sr-xr-x   1 root     sys        36456 Sep  1  1998 ps
-r-xr-xr-x   1 bin      bin        17168 May 20  1999 pvs
-r-xr-xr-x   1 bin      bin        16360 Jun 29  1999 savecore
-r-xr-xr-x   1 bin      bin        11744 Sep  1  1998 setuname
-r-xr-xr-x   1 bin      bin        11840 Sep  1  1998 tnfxtract
-r-xr-xr-x   1 bin      bin       178408 Jun 29  1999 truss
-r-sr-xr-x   2 root     bin        15584 Sep  1  1998 uptime
-r-sr-xr-x   2 root     bin        15584 Sep  1  1998 w
#


Anyone know where I can find a 64 bits version of find ?

Frangois Bousquet
Counselor - Unix System Administrator
INET - Internet Maintenance
CGI - LBC Services
Email : Francois.Bousquet at CGI.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From Laurence_Cohen at sra.com  Wed Jan 26 14:40:24 2005
From: Laurence_Cohen at sra.com (Cohen, Laurence)
Date: Wed, 26 Jan 2005 14:40:24 -0500
Subject: SUMMARY: Inaccurate file size for large files
Message-ID: <5C92F9725F58D34791C636335A4A575501A05FAD@flex1.sra.com>

Very quick response!  Basically Mike Salehi and Tom Schmidt indicated that I
am probably seeing a sparse file which has a larger address space than the
actual size of the file.  According to Mike, this can happen with database
files, which this file happens to be, and this isn't a problem.

Thanks to everyone who responded!

>  -----Original Message-----
> From: 	Cohen, Laurence
> Sent:	Wednesday, January 26, 2005 2:27 PM
> To:	'sunmanagers at sunmanagers.org'
> Subject:	Inaccurate file size for large files
>
> Hello everyone,
>
> I tracked down a file on one of our file systems that is indicating it is
over 350GB in size.  This obviously makes no sense, as I believe the
limitation of a file is 32GB.  Has anyone heard of a bug that causing the file
size to be misrepresented when it reaches 32GB?  I looked on SunSolve but
could not find anything.
>
> Thanks!
>
> Laurence H. Cohen
> System Administrator
>
>
>
>
> --
> No virus found in this outgoing message.
> Checked by AVG Anti-Virus.
> Version: 7.0.300 / Virus Database: 265.7.4 - Release Date: 1/25/2005
>
>
--
No virus found in this outgoing message.
Checked by AVG Anti-Virus.
Version: 7.0.300 / Virus Database: 265.7.4 - Release Date: 1/25/2005
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From cadolf at sandia.gov  Wed Jan 26 18:29:43 2005
From: cadolf at sandia.gov (Adolf, Christopher)
Date: Wed, 26 Jan 2005 16:29:43 -0700
Subject: SUMMARY: Korn Shell Execution in verbose mode
Message-ID: <AB4680A899D2F441A43D31F2A9B8451B210799@ES22SNLNT.srn.sandia.gov>

I wanted to thank everyone who offered suggestions.  Because of them I was
able to see what I needed to.

So, quick summary:

I need to parse through a script to get the exact syntax of a command it is
running.  Unfortunately the script is written with functions.

ksh -vx will give me the output processing of the scripts, but doesn't give
the output processing of the functions inside the script.  Does anyone know a
way to tell it to give that information to me as well.

Solutions:

I can place a "set -vx" internal to the functions.  This approach worked
wonderfully!
Someone also suggested there is a version of ksh made for debugging that you
can possibility get from the OS vendors.

Thanks again for everyones help!
Chris
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From zoo11 at hst.co.kr  Wed Jan 26 19:57:17 2005
From: zoo11 at hst.co.kr (Joohyun Cha)
Date: Thu, 27 Jan 2005 09:57:17 +0900
Subject: SUMMARY: Moving disk slice.
Message-ID: <20050127005717.GA24466@mail.hst.co.kr>

Hi. Managers.

Thanks for your kind reply.

Nearly all admin. said that I can't do it without rebuilding slice table and
file system. 

but hhuseman at aragorn.dh9dat.de gave me mail like this.

---
two questions before:
What OS is the machine running?? Solaris 8 or greater? If yes, there IS a
+possibility to solve this - if no, sorry, no idea...
Do you have some place to copy away root.back and home.back? This one's have to
+be empty...

If the answer to both questions is "YES", you can solve the problem:

Install "Solstice DiskSuite" (if it's Sol 9, SDS is part of the OS, and called
+"LVM" - but, it's exactly the same stuff, :-) ).
Then, create a one-way-mirror out of user.back, and reboot the machine to mount
+it. Afterwards, add the other two slices as additional
stripes, and allow SDS to do a "growfs". That should do the trick, I did it very
+often in the past.
The best tool for the SDS-stuff is the GUI, called metatool. Just take a look in
+the documentation, metatool is quite nice, :-))
---
Although OS isn't Solaris 8 (it's 7), I could do it with Disksuit 4.2.1
without any problem. Here's the way I got...

1) backup /root.back and /home.back
2) install SDS 4.2.1/patches and reboot.
3) delete slice s0, s5 of c0t1d0 and make three small (4MB) slice to 
   maintain metadb. make another big slice to attach s6 after doing SDS-stuff.
4) metadb init 
   # metadb -a -f /dev/rdsk/c0t1d0s0 
   # metadb -a /dev/rdsk/c0t1d0s1
   # metadb -a /dev/rdsk/c0t1d0s3
5) make s5 and existing s6 metadevice. (concat)
   # metainit d10 2 1 c0t1d0s6 1 c0t1d0s5
   Device order is important.
6) grow its filesystem
   # growfs /dev/md/rdsk/d10
7) mount/test it.

Thanks all!

My origianl query....
---
Hi. Managers.

I have disk that has 3 slices in it and is growing data
size rapidly. Here's df -k output and slice table.

/dev/dsk/c0t1d0s0     288603   42112  217631    17%    /root.back
/dev/dsk/c0t1d0s5    2056211  184984 1809541    10%    /home.back
/dev/dsk/c0t1d0s6    15042643 14141316  750901    95%    /user.back

Part      Tag    Flag     Cylinders        Size            Blocks
  0       root    wm       0 -  130      301.40MB    (131/0/0)    617272
  1 unassigned    wu       0               0         (0/0/0)           0
  2     backup    wm       0 - 7505       16.86GB    (7506/0/0) 35368272
  3 unassigned    wm       0               0         (0/0/0)           0
  4 unassigned    wm       0               0         (0/0/0)           0
  5 unassigned    wm     131 - 1021        2.00GB    (891/0/0)   4198392
  6 unassigned    wm    1022 - 7505       14.57GB    (6484/0/0) 30552608
  7 unassigned    wm       0               0         (0/0/0)           0

I want to make big one slice by adding all 3 slices so that /user.back
can contain more data.

Is there any method for this except rebuild whole slice table and file
system?

Thanks in advance!
---

-- 
------------------------------------------------------------------------------
From:
main(){int a=122,j=11;while(a>-50){a=a>0?a:111;printf("%c",a);a=j==49?46:a-j;
j*=j;j=j%2?j:7;j*=j%8?1:0;}printf("%c1\n",49);}
------------------------------------------------------------------------------
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From dan.cave at gmail.com  Thu Jan 27 05:40:24 2005
From: dan.cave at gmail.com (Dan Cave)
Date: Thu, 27 Jan 2005 10:40:24 +0000
Subject: SUMMARY: JumpstartWebflash Images :
Message-ID: <2731bd3b05012702403ef9ce98@mail.gmail.com>

A belated thanks to Iain Millar, Daniel Nuqo and Dan O'Callaghan who
kindly enlightend me to use the -S switch which doesn't spend time
processing the flar archive size, along with some very useful
information. I was also told that when recovering the flash images,
the jumpstart server decompresses the image into swap/memory.

I managed to reduce the time from 2hr, 34mins -> 1hrs 35m just by
using the -S switch., That was 30 minutes faster than using a
traditonal dds4 tapedrive on scsi3.

Interestingly, the time it took to archive the 22Gb flash archive over
1Gb network directly to the sagita san from the client over nfs was
exactly the same time writing to the JBOD raid 5 disk over Gbit NFS
which was attached at the jumpstart server. (i'd be interested to hear
you comments). Possibly a limiting factor of NFS?

Also, thanks to those of you, including Scott Cooper, who helped out
with getting me sorted with my original flasharchive/jumpstart server
and solaris2.6 questions.

You know who you are, thanks!!!

regards

dan.

My original post below with Dan's reply.

---------- Forwarded message ----------
From: Dan O'Callaghan
Date: Tue, 25 Jan 2005 10:05:34 +0000
Subject: Re: Jumpstart webflash image resore times.
To: Dan Cave <dan.cave at gmail.com>


Dan
You can save time with flarcreate by using the "-S" switch, stops it
calculating the size.

I've never had flarcreate accurately calculate the image anyway, i think
(could be very wrong here) but for this to work properly you need as
much swap space as the size of the flar.
The boxes i create are usually about 2-3gig in size and when i create
the initial image, the swap is usually around 500M. I've noticed that is
spends about an hour trying to calculate the size of the flar and then
gives up and creates the image as if i had uses the "-S" switch.
When you restore the image, i always have the message "no size
information of the flar archive". I'm using Solaris 8 2/04.
One day i'll add a temp swap file to see if i'm right.

On Thu, 2005-01-20 at 15:54 +0000, Dan Cave wrote:
> Hi all.,
>
> Previous to my last two emails, I thought I'd update you and ask for a
> bit of feedback.
>
> My orginal email asked if anyone knew whether it was possible to
> backup and restore solaris2.6 images created via flarcreate and
> restored via a jumpstart webflash archive on a solaris 8 box.
>
> Well, I successfully managed to build a jumpstart server (sol8) with
> flash image capability, recover the images after using Gbit ethernet
> point to point, and copying the flarcreate script to the 2.6 box,
> creating a 26Gb flar image(via a local nfs mount), which we usually
> store on a dds3/4 tape using ufsdump/ufsrestore.
>
> However, our usual tape resore takes just over two hours from a home
> rolled restore cd from which we boot.
>
> The restored flar image from jumpstart takes over two hours also,
> however I would have expected it to be quicker as I'm using Gbit
> ethernet.
>
> Has anyone had any experiences with restores from flar images which
> take roughtly the same ammount of time from tape? I've been
> enlightened that you cannot use compressed flash archives as they are
> purely for storage purposes.
>
> I was really looking to save time when creating and  restoring flar
> (FRU's as sun call them) images using jumpstart, but I'm a little
> bemused as to why this is so.... I had applied the 2Gb patch for
> restoring flar images and have a working system from which I created
> my 26Gb image. (a mix of OS data, applications and user data)
>
> It takes 2hrs 43 minutes to work out and create my flar image over a
> Gbit net, as I also believe that roughly 40% of this time is spent
> working out how large the image is going to be. Some editing of the
> script and adding in timing debug which I added in and created a flar
> archive of my jumpstart server to work this out.
>
> Has anyone else experienced similar behaviour?
>
> Feedback is most definitely appreciated. I shall summarise..
>
> Regards
>
> dan.
> _______________________________________________
> 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 gary.paveza at AIG.COM  Thu Jan 27 10:12:10 2005
From: gary.paveza at AIG.COM (Paveza, Gary)
Date: Thu, 27 Jan 2005 10:12:10 -0500
Subject: SUMMARY:  DR / cloning Solaris
Message-ID: <79CF2118BE3ED511AA110003476BCB9405327120@xwilmpf03.aig.com>

Thanks to everyone who replied.  Most told me that flash'ing a system with
metadrives is not a problem, you just need to recreate them after the
install.  For my situation that's all I need (all user data is recovered via
backups after the system is installed).

---------------------------------------------------------
Gary Paveza, Jr.
Senior Systems Administrator -CSA
(302) 252-4831 - phone


		-----Original Message-----
		From: Paveza, Gary 
		Sent: Thursday, January 20, 2005 8:24 AM
		To: 'sunmanagers at sunmanagers.org'
		Subject: DR / cloning Solaris

		What are people using for disaster recovery / cloning of
their solaris servers?  

		I was looking into Jumpstart, but that appears to be for
cold installation from CD images.

		I also looked at Flash archives, but that appears not to
support metadevices.  

		I'm running both Solaris 8 and 9.
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From jimbeyyl at yahoo.com  Thu Jan 27 15:24:55 2005
From: jimbeyyl at yahoo.com (jimi beyylorn)
Date: Thu, 27 Jan 2005 12:24:55 -0800 (PST)
Subject: SUMMARY: problem  install openssl
Message-ID: <20050127202456.95569.qmail@web30701.mail.mud.yahoo.com>

Thanks all, i had good reply from: 
 
"David Foster" foster at ncmir.ucsd.edu
"Bousquet Francois" Francois.Bousquet at cgi.com
"Geoff Wright" geoff at xp3.co.uk
"Crist Clark" crist.clark at globalstar.com
"Geoffrey Walton" setuid0 at mac.com 
 
I get OpenSSL from http://www.sunfreeware.com
 
--------------------------------------------------------------------------------
Hi all,

While trying to install v0.9.7e of openssl on a Solaris 9 machines , 
the build completes fine, but it bombs out during make install and 
display followin error :
 
installing fips... making install in fips/sha1... sh: syntax error at 
line 2: `;' unexpected *** Error code 2 make: Fatal error: Command 
failed for target `install' Current working directory 
/usr/local/openssl/fips/sha1 *** Error code 1 make: Fatal error: Command failed for target 
`install' Current working directory /usr/local/openssl/fips *** Error 
code 1 make: Fatal error: Command failed for target `install_sw'
 
The compiler is the Sun Workshop compiler and make as installed in
/usr/ccs/bin/make.

Regards
jimi
 Yahoo! Search presents - Jib Jab's 'Second Term'
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From sguillemette at metrocastcablevision.com  Thu Jan 27 15:37:34 2005
From: sguillemette at metrocastcablevision.com (Shawn Guillemette)
Date: Thu, 27 Jan 2005 15:37:34 -0500
Subject: Summary : NFS Mounts
Message-ID: <41F9510E.6000503@metrocastcablevision.com>

This system I spoke of was set up by another party. I have done some 
research internally and discovered that the panic was a false alarm. The 
mounts I was looking for were not really mounts to the netapp. The MX 
that had seemed to be fine is a much beefer box it has its on disk to 
mount. the rest are with in the system mounts.

I thank you for you time and suggestions.

Sorry to lead you down the wrong path.


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


From crussell_1969 at yahoo.com  Thu Jan 27 16:18:32 2005
From: crussell_1969 at yahoo.com (Craig Russell)
Date: Thu, 27 Jan 2005 13:18:32 -0800 (PST)
Subject: Summary:sed question
Message-ID: <20050127211832.94958.qmail@web41212.mail.yahoo.com>

Many thanks-

Most people pointed out that I was using single quotes
around my sed command which was preventing the shell
from expanding the variables inside which was true for
the incantation I put in the question.  I ended up
using a different solution recommended by 

arno uhlmelmahei" <arno at panix.com>

sed -n 1,33p file >newfile
     cat datafile >>newfile
     sed -n '43,$p' file >>newfile

I placed this into a for loop and got the result I was
looking for.

Many thanks to the following for their suggestions:

JULIAN, JOHN C (AIT)" <jj2195 at sbc.com
francisco" <frisco at blackant.net
Kurt Schulte" <r2kurts at yahoo.com>
Ray Brownrigg" <ray at mcs.vuw.ac.nz
Eric Noriega" <noriega at gwu.edu
John Leadeham" <jl at lammtarra.fslife.co.uk


Also, thanks to John Leadeham for steering me towards
the usenet group news:comp.unix.shell for similar
questions.

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


From sun1sol at yahoo.com  Thu Jan 27 17:46:22 2005
From: sun1sol at yahoo.com (ert weerr)
Date: Thu, 27 Jan 2005 14:46:22 -0800 (PST)
Subject: SUMMARY: Online log file monitoring script?
Message-ID: <20050127224623.67197.qmail@web52209.mail.yahoo.com>

Guys,

> I have to write a script that online monitors a
> log file to check if a particular error message
> turns up there.
> The log file is quite big and I need to run some
> commands right after the message appears in the log.
> So, to grep for the message in every 1 second won't
> work, because at the time the script finished
> to grep the huge file some valuable seconds 
> would elapse.

Many thanks to everyone for your invaluable help!!!
The world would be a harsh place without you guys!

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

You can look at swatch and/or logcheck (or is it
checklog).
swatch is more for continual monitoring (taking more 
immediate action), logcheck more for periodic. 
logcheck tracks where it left off in the file and
seeks to that point before continuing, so it may have 
the performance requirements you need.

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

logsurfer, logcheker, etc.

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

tail -f logfile | grep pattern |
while read line; do
	echo found it: $line
	#do other stuff
done

Run that as a daemon process. -f

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

logbot
logsurfer
swatch
syslog-ng
jabber

and whatnot on freshmeat.net or sunfreeware.com. They
all have this similar functionality.

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

   #!/bin/ksh

   tail -f logfile |
   while read LINE; do
	if echo $LINE | grep 'HELLO' >/dev/null; then
		I=$(($I + 1))
		echo "Encountered $I HELLOs"
	fi
   done

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

#!/bin/sh
tail -f /yourlog|grep yourmsg|while read line; do
echo execute commd using $line
done

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

http://sourceforge.net/projects/simple-evcorr/
http://www2.logwatch.org:81/

There are others...

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

If you're thinking of perl, the File::Tail module can
help you out here. 
If not, you can still do something like 
tail -f <filename> | grep 
"mystring" in you script, but you'll have to watch out
for logfile 
rotation etc.

Regards,

John


		
__________________________________ 
Do you Yahoo!? 
Yahoo! Mail - now with 250MB free storage. Learn more.
http://info.mail.yahoo.com/mail_250
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From omon.edeki at nowellgroup.com  Fri Jan 28 03:20:57 2005
From: omon.edeki at nowellgroup.com (Omon Edeki)
Date: Fri, 28 Jan 2005 03:20:57 -0500
Subject: SUMMARY Retrieving TCP /UDP Table, Netstat source code
Message-ID: <05Jan28.032105-0500_est.372071-25447+25916@ams003.ftl.affinity.com>

Thanks to 

1.Scott Mickey
2.Eric Noriega
3.Neil Quiogue
4.Nelson Bensley 

and others who contributed. raw TCP/UDP table information can be retrieved
from the kstat library.. 

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


From kdavidson at transaction-one.com  Fri Jan 28 03:57:50 2005
From: kdavidson at transaction-one.com (Kevin Davidson)
Date: Fri, 28 Jan 2005 08:57:50 -0000
Subject: SUMMARY: DLT8000 not unmounting tape
In-Reply-To: <1106825504_359@cheech>
Message-ID: <1106902494_1480@cheech>

Thanks to everybody who replied. Suggestions mainly focused on either the
drive or the media being bad and needing replacement. However, after a two
hour trip across the city to the datacentre, the problem turned out to be
human error on the part of the NOC staff responsible for changing the tapes!


Thanks to 

Ian Pease
John Christian
Rodrigo Javier Hernandez Egnen
Angelo Bonfieti Junior
Rafael Hinojosa
Joe Fletcher

For their prompt replies and suggestions.

Kevin

-- 
Kevin Davidson  |       UNIX Systems Administrator
Transaction One |       www.transaction-one.com
020 8358 9126   |       kdavidson at transaction-one.com
 
-----Original Message-----
From: sunmanagers-bounces at sunmanagers.org
[mailto:sunmanagers-bounces at sunmanagers.org] On Behalf Of Kevin Davidson
Sent: 27 January 2005 11:35
To: sunmanagers at sunmanagers.org
Subject: DLT8000 not unmounting tape

Hi everybody,

I have a DLT8000 which won't unmount the tape that is currently loaded.

According to the OS there is no tape mounted:-

$ mt -f /dev/rmt/0 status
/dev/rmt/0: no tape loaded or drive offline

I have the following error in /var/adm/messages

WARNING: /pci at 1f,4000/scsi at 3,1/st at 4,0 (st11):
Error for Command: write                   Error Level: Fatal
Requested Block: 716224                    Error Block: 716224
Vendor: QUANTUM                            Serial Number:  P  "
Sense Key: Media Error
ASC: 0xc (write error), ASCQ: 0x0, FRU: 0x0

It looks like the backup last night filled the tape.

Any ideas, or should the drive be replaced?

Thanks in advance, will summarize the responses.

Kevin

-- 
Kevin Davidson  |       UNIX Systems Administrator
Transaction One |       www.transaction-one.com
020 8358 9126   |       kdavidson at transaction-one.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From rob.de.langhe at belgacom.be  Fri Jan 28 07:39:35 2005
From: rob.de.langhe at belgacom.be (rob.de.langhe at belgacom.be)
Date: Fri, 28 Jan 2005 13:39:35 +0100
Subject: SUMMARY on non-posted question: net- boot hangs with
 non-standard netmasks
Message-ID: <20E53B61AAB3994EA15681023EE52418DFB246@AE0007.BGC.NET>

Hi,

just want to share this one with you, since I searched the net on
similar symptoms but did not find any related discussion:

Our Jumpstart setup is :

client boots via DHCP from central DHCP servers, using "DHCP-relaying"
by the inter-segment routers.

The DHCP-servers return boot-parameters to the client, informing it
about what IP-address to use, netmask, default gateway, and from which
TFTP-server to load the initial boot-image. So far so good : the client
configures its interface with that information, and accesses the
TFTP-server on a remote subnet using the correct netmask and default
gateway.

Once the boot-image is loaded, it requests again via DHCP where to find
the mini-Solaris O/S that is "Jumpstart-aware". Reply is correctly send
back, and the client NFS-mounts correctly that "BOOT-Solaris" from the
install-server which happens to be on a remote subnet (but the client
still uses the correct netmask and default gateway).

Once the client boots with that mini-Solaris, it will again configure
its interfaces in the script "/sbin/rcS". It does so by calling a binary
"/sbin/get_netmask" with the IP-address of the NFS-server. This program
returns the netmask of the NFS-server, apparently. But that netmask is
NOT the mask that the client should use on its own segment.

So the client than proceeds to attempt an NFS-mount of the same
install-server, to find the "install" bundles. But since it has set an
incorrect netmask on its (previously correctly configured) interface, it
cannot reach the NFS-server anymore. Result : the boot hangs with

...(booting)
Configured interface xyz
NFS server 1.2.3.4 not responding still trying
NFS server 1.2.3.4 not responding still trying
... (endlessly)


**** DISCLAIMER ****
http://www.belgacom.be/maildisclaimer
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From sloewenthal at gemini.edu  Fri Jan 28 09:29:39 2005
From: sloewenthal at gemini.edu (LOEWENTHAL Simon)
Date: Fri, 28 Jan 2005 11:29:39 -0300
Subject: SUMMARY - SUNWqus QLogic Ultra3 Scsi - Need revision 11.9.0.
In-Reply-To: <BDD35658B859D51187AB00508BB8203005D2C75F@nswcdlvaex12.nswc.navy.mil>
Message-ID: <DOEBKMNALLLEDCCBEFPNKENEDHAA.sloewenthal@gemini.edu>

Thanks to Clift Roberts Chris Lovett John Leadeham Andrew Watkins and Sam
Impastato for pointing me towards the right places.

http://www.sun.com/download/index.jsp?cat=Systems%20Administration&tab=3&sub
cat=Storage%20Management

---
ISG/Gemini South/AURA
+56.51.205610
www.gemini.edu

-----Original Message-----
From: Clift Robert T CONT DLVA [mailto:CliftRT at NSWC.NAVY.MIL]
Sent: Friday, January 28, 2005 10:58 AM
To: 'LOEWENTHAL Simon'
Subject: RE: SUNWqus QLogic Ultra3 Scsi - Need revision 11.9.0.



--
No virus found in this outgoing message.
Checked by AVG Anti-Virus.
Version: 7.0.300 / Virus Database: 265.8.1 - Release Date: 27/01/2005
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From rob.de.langhe at belgacom.be  Fri Jan 28 07:45:24 2005
From: rob.de.langhe at belgacom.be (rob.de.langhe at belgacom.be)
Date: Fri, 28 Jan 2005 13:45:24 +0100
Subject: SUMMARY on non-posted question: net- boot hangs with
 non-standard netmasks
Message-ID: <20E53B61AAB3994EA15681023EE52418DFB25A@AE0007.BGC.NET>

forget to mention the workaround :

on the install-server, move aside that program
".../Boot/sbin/get_netmask" (mv get_netmask get_netmask.nouse) so that
the client does not change its netmask.
That's all...

-----Original Message-----
From: sunmanagers-bounces at sunmanagers.org
[mailto:sunmanagers-bounces at sunmanagers.org] On Behalf Of
rob.de.langhe at belgacom.be
Sent: 28 January 2005 13:40
To: sunmanagers at sunmanagers.org
Subject: SUMMARY on non-posted question: net- boot hangs with
non-standard netmasks

Hi,

just want to share this one with you, since I searched the net on
similar symptoms but did not find any related discussion:

Our Jumpstart setup is :

client boots via DHCP from central DHCP servers, using "DHCP-relaying"
by the inter-segment routers.

The DHCP-servers return boot-parameters to the client, informing it
about what IP-address to use, netmask, default gateway, and from which
TFTP-server to load the initial boot-image. So far so good : the client
configures its interface with that information, and accesses the
TFTP-server on a remote subnet using the correct netmask and default
gateway.

Once the boot-image is loaded, it requests again via DHCP where to find
the mini-Solaris O/S that is "Jumpstart-aware". Reply is correctly send
back, and the client NFS-mounts correctly that "BOOT-Solaris" from the
install-server which happens to be on a remote subnet (but the client
still uses the correct netmask and default gateway).

Once the client boots with that mini-Solaris, it will again configure
its interfaces in the script "/sbin/rcS". It does so by calling a binary
"/sbin/get_netmask" with the IP-address of the NFS-server. This program
returns the netmask of the NFS-server, apparently. But that netmask is
NOT the mask that the client should use on its own segment.

So the client than proceeds to attempt an NFS-mount of the same
install-server, to find the "install" bundles. But since it has set an
incorrect netmask on its (previously correctly configured) interface, it
cannot reach the NFS-server anymore. Result : the boot hangs with

...(booting)
Configured interface xyz
NFS server 1.2.3.4 not responding still trying NFS server 1.2.3.4 not
responding still trying ... (endlessly)


**** DISCLAIMER ****
http://www.belgacom.be/maildisclaimer
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


**** DISCLAIMER ****
http://www.belgacom.be/maildisclaimer
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From bruce.truax at ge.com  Fri Jan 28 17:24:00 2005
From: bruce.truax at ge.com (Truax, Bruce (GE Commercial Finance, NonGE))
Date: Fri, 28 Jan 2005 17:24:00 -0500
Subject: Summary: A1000 lad issue
Message-ID: <352EBFE67A6D3543B12C37B2CB74A57F0A9EFF@CINMLVEM08.e2k.ad.ge.com>

Sunmanagers,

I have a solution to my A1000 lad issue. The original message was:

		I have an A1000 exhibiting odd behavior. It appears to be operating
correctly except for when I go to view the luns using lad. Lad reports 10, yes
10, identical instances of the controller. Since the machine is in a test
environment, I have been able to connect it to a "clean" machine and I get the
same results. Has anyone seen this before and is it caused by the controller?
I have tried a different, known good terminator and cable to eliminate any
"echo" on the SCSI bus.

I received 3 responses, 2 of which had me verify the connection to the server
and validating the terminator's condition. The 3rd response from  David Deaves
solved my issue. His response read:

		The only times I have seen things occur multiple times on a scsi bus is when
their scsi id matched the controller's (normally 7), so maybe check that.

After checking the ID, I discovered I had overlooked this and Dave was right.

Again, Thanks for the help.
Bruce
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From crist.clark at globalstar.com  Fri Jan 28 17:54:46 2005
From: crist.clark at globalstar.com (Crist Clark)
Date: Fri, 28 Jan 2005 14:54:46 -0800
Subject: SUMMARY: Setting Up iPlanet Directory 5.1
In-Reply-To: <41F02A97.6010604@globalstar.com>
References: <41F02A97.6010604@globalstar.com>
Message-ID: <41FAC2B6.1070801@globalstar.com>

Thanks to a number of people who simply pointed me to,

	/usr/sbin/directoryserver setup

And the directoryserver(1M) manpage. It'd be nice if the iPlanet
manuals would mention that. However, once I got going with that, I
have to say it was pretty easy to setup.

One other small issue I ran into. I ended up configuring iPlanet LDAP
on two servers. The one with patch 113859-02 had troubles starting up
the administration server. A plugin in the magnus.conf caused dynamic
linking problems. I deleted the plugin since it wasn't present in the
other magnus.conf and the admin server on that one seemed to be OK.
The admin server would then start up without a problem. The library
it couldn't find was libjvm.so. There are a number of those sprinkled
around the system. I probably could have fixed this by getting the
right hints in LD_LIBRARY_PATH, but rather than try to figure out
which one I should use, I just dropped the plugin.

Crist Clark wrote:

> I'm having some trouble setting up iPlanet Directory Server 5.1. The
> packages are already installed on the system (IPLTdsr, IPLTdsu, IPLTdscon,
> IPLTdsman), but no directory server has been configured. I've been trying
> to follow the installation instructions at docs.sun.com, but they all
> seem to assume you are installing fresh from a tarball.
> 
> I found what I thought was the setup program I needed, in,
> 
>   /usr/iplanet/ds5/setup/setup
> 
> Things seem to go well, until it tries to extract the files for
> installation from the zip file. There is no zip file, the packages
> are already installed.
> 
> Anyone have pointers on how to get a directory server up and running
> _after_ the packages have been installed?


-- 
Crist J. Clark                               crist.clark at globalstar.com
Globalstar Communications                                (408) 933-4387

The information contained in this e-mail message is confidential,
intended only for the use of the individual or entity named above.
If the reader of this e-mail is not the intended recipient, or the
employee or agent responsible to deliver it to the intended recipient,
you are hereby notified that any review, dissemination, distribution or
copying of this communication is strictly prohibited.  If you have
received this e-mail in error, please contact postmaster at globalstar.com
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From stoyan.genov at sun-fish.com  Mon Jan 31 08:44:09 2005
From: stoyan.genov at sun-fish.com (Stoyan Genov)
Date: Mon, 31 Jan 2005 14:44:09 +0100
Subject: SUMMARY: 2xV880 + 2xT3+ + 2xFC-switch in a redundant
 configuration: what types of cables?
In-Reply-To: <41F14076.3020304@sun-fish.com>
References: <41F14076.3020304@sun-fish.com>
Message-ID: <41FE3629.9020302@sun-fish.com>

In short:
*) X6729A is not supported on V880, use X6799A instead;
*) To connect X6799A HBA to the switch, use an SC-LC cable;
*) To connect a T3+ to the switch, use an LC-LC cable;
*) Use an SFP to connect the "LC-end" of the cable to the switch;

Million thanks to Drew Skinner, David Livingston and E. (NO UCE)
for the following:

Drew Skinner pointed out that the master of a T3+ partner pair
would shut down both arrays if you fail to replace a FRU (a disk)
within a certain time frame, and that in his opinion it's better if
the T3+s are mirrored using SDS or similar software.

David Livingston pointed out that I will need SC-LC cables to connect
the HBAs to the switches, LC-LC cables to connect the T3+s to the
switches, and SFP (370-5211) modules to connect the LC-end of the cables
to the switches.

E. (quote) "throwed a spanner in the works" by pointing out that X6729A
isn't supported on USIII or higher CPU systems (this includes V880).
Actually, this saved me a lot of grief, thanks, E!. Also, confirmed what
cables I need.

A great list, this is!

Regards,
Stoyan Genov

Stoyan Genov wrote:
> Good day,
> 
> I have to set a pair of Sun Fire V880 machines, a pair of T3+ storage
> arrays and a pair of SG-XSW8-2GB FC switches in the following way
> (the idea of the setup being high availability of the data on the T3+s):
> 
> [V880]---[fc switch]---[T3+]
>    |         |           |
>    +--------------+      |
>              |    |      |
>    +---------+    |      |
>    |              |      |
> [V880]---[fc switch]---[T3+]
> 
> (ASCII art, please use a fixed width font to see drawing correctly)
> 
> Each V880 has a pair of X6729A FC HBAs, and each of the X7629A's
> connects to a port of a SG-XSW8-2GB switch.
> 
> Each T3+ connects to a port of a switch (each array to different switch)
> 
> The two T3+s are interconnected in a partner group.
> 
> I need to obtain cables for all connections.
> The question is, what exactly types of cables do I need?
> Here are some thoughts on the subject:
> 
> 1. T3+ interconnect cable: no variations. Just one type: 530-2843
> 
> 2. In the specs of the switch(es)
> (http://sunsolve.sun.com/handbook_pub/Devices/Fibre_Channel/FIBRE_2GB_FC_Switch_8.html) 
> 
> I see that all cables are either LC-LC or LC-SC, which makes me think
> I need LC on the switch's side. However, notes 3 and 4 on the same page
> bother me -- I do not know whether the switches are 370-5211 or 370-5212
> 
> 3. In the specs of the T3+
> (http://sunsolve.sun.com/handbook_pub/Devices/Fibre_Channel/FIBRE_T3_RAID_Ctlr.html#5710) 
> 
> I see all cables are LC-SC, and there's an SC coupler offered.
> Does that mean that I have to use an LC-LC cable to connect a T3 to
> a switch, or an LC-SC+coupler+SC-LC? Or I have completely messed up?
> 
> 4. In the specs of the HBA
> (http://sunsolve.sun.com/handbook_pub/Devices/Fibre_Channel/FIBRE_Single_FC_AL.html) 
> 
> I can see the connector is SC duplex; does that mean I have to use an
> SC-LC cable to connect a HBA to a switch?
> 
> 
> Any comments, ideas and answers on the setup and the cables
> will be greatly appretiated. I will summarize.
> 
> Best Regards,
> Stoyan Genov
> _______________________________________________
> 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 Jeremy.Ahl at vivista.co.uk  Mon Jan 31 10:58:13 2005
From: Jeremy.Ahl at vivista.co.uk (Jeremy Ahl)
Date: Mon, 31 Jan 2005 15:58:13 -0000
Subject: SUMMARY: RSC console hangs
Message-ID: <E671455BAF0BFC43B8B31A6455348E7F06FA0784@srvmeth10.sis.co.uk>

Hello all,

Apologies for the late summary, but it seems that this problem was related
to being logged in as root via the console at the same time.
If/when you logout of the rsc, you get no prompt back until you also log out
your console session.

Jez
>  -----Original Message-----
> From: 	Jeremy Ahl  
> Sent:	08 March 2004 10:13
> To:	'sunmanagers at sunmanagers.org'
> Subject:	RSC console hangs
> 
> Hi
> 
> Apologies if this has been asked before, but I haven't seen it anywhere.
> 
> V880 using RSC 2.2.
> 
> When I telnet to the rsc, and type console nothing happens. ~. returns me
> to the rsc prompt.
> The only thing that fixes it is to reset the V880. It seems to then be OK
> for the first time you use it, then hangs again.
> RSC is patched to 111500-09
> 
> Thanks in advance
> 
> Jez



_____________________________________________________________________

This message has been checked for all known viruses on behalf of Vivista by MessageLabs. 

http://www.messagelabs.com or Email: mailsweeper.info at vivista.co.uk

Vivista formerly Securicor Information Systems for further information http://www.vivista.co.uk  
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers


From cbar44 at tsg.cbot.com  Mon Jan 31 14:04:10 2005
From: cbar44 at tsg.cbot.com (Christopher L. Barnard)
Date: Mon, 31 Jan 2005 13:04:10 -0600 (CST)
Subject: SUMMARY: simple regex question
In-Reply-To: <Pine.GSO.4.58.0501281852520.1273@shalmaneser.tsg.cbot.com>
References: <Pine.GSO.4.58.0501281852520.1273@shalmaneser.tsg.cbot.com>
Message-ID: <Pine.GSO.4.58.0501311302550.1273@shalmaneser.tsg.cbot.com>

I asked:

> I have a quick regexp question.
>
> I want to match, case does not matter, a string
> foo: bar
> where foo starts the line, there can be one or more whitespace characters
> between the : and the bar, and there can be zero or more whitespace
> characters between the bar and the end of line.  "whitespace characters"
> here means spaces or tabs.
>
> /usr/bin/egrep -i ^foo:[ ][ ]*bar[ ]*$
>
> works with any combination of spaces I use.  But when I add \t to the regex:
>
> /usr/bin/egrep -i ^foo:[ \t][ \t]*bar[ \t]*&
>
> it fails.  I have tried escaping the \t (as in, \\t), but that did not help.
> I tried bracketing with ( and ), but that did not help.  I have used double
> and single quotes, and that did not help.  Can anyone suggest how to write an
> RE that will see a tab character?  TIA and I will summarize.

The answer:

The "\t" is how it is written down in documentation.  But to use the tab
character in a regex, hit the tab key.  literally.  It works like a charm.

For the individuals who suggested /:space:/ : yes, that would work in a more
powerful shell.  I want this to work in the basic /bin/sh, and understanding
of /:space:/ in egrep is something added in ksh.

For the individuals who did not understand why I would have the [ =t]*$ at
the end: that will capture whitespace between what looks like the last
character of the string and the end-of-line.  Whitespace at the end doesn't
harm anything, so there is no reason to check for it.

I also got a few links to get the hex code for a tab.  Thanks.  (TAB= \011).

Thanks to:

Doc G <maddocg at verizon.net>
Alex Stade <alex at trdlnk.com>
Larry Anta <lanta at ryerson.ca>
Kurt Schulte <r2kurts at yahoo.com>
Darren Dunham <ddunham at taos.com>
Karyn Williams <karyn at calarts.edu>
Colin Bigam <colin at west.gecems.com>
Crist Clark <crist.clark at globalstar.com>
John Leadeham <jl at lammtarra.fslife.co.uk>
David Markowitz <DMarkowitz at litronic.com>
Hans Jacobsen <HJacobsen at grandcentral.com>
R A Lichtensteiger <sunmanagers at throwawaydomain.com>

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


From tgrassia at sfnewmexican.com  Mon Jan 31 18:35:29 2005
From: tgrassia at sfnewmexican.com (Tom Grassia)
Date: Mon, 31 Jan 2005 16:35:29 -0700
Subject: SUMMARY:  Attaching a D1000 to a U2
Message-ID: <6.1.2.0.0.20050131092810.025e9278@mail>

Thank you all for your help.  We laughed, we cried, we had a grand time.

The overwhelming response mentioned boot -r.  I had though this odd, since 
my experience in adding new drives had been one of two things..

#1.  Power down machine.

#2.  Add new drives.

#3.  When machine comes up, do reboot -r and configure drives when system 
comes back.

OR

#1 touch /reconfigure before powering down machine.

Since the system started to error out before I reconfigured the devices, I 
was at a loss on how that would fix the issue.  However, I had nothing 
better to do, so I synthesized a number of suggestions into one 
procedure.  The procedure is as follows:


1. sync;sync;init0

2. from the ok prompt:
setenv auto-boot? false
reset-all or resetall
probe-scsi-all

See the current disks come up, make notes on location.

3.  From some emails, I have received the impression that some people add 
hardware while the system is powered on and in single user mode.  I power 
down the system, as that is the way of my father and his father before him.

poweroff

4.  Attached the D1000's SCSI cable into the differential SCSI card which 
was still in the server.

5. Pray

6. Power the system back on.  At the ok prompt I checked the scsi chains 
again...
probe-scsi-all

All the drives appeared to be in the same locations, + there are now six 
new drives!

7. boot -r

The system reconfigures the drives and comes back up.  I can now access and 
configure the new drives in the D1000.

I'm pretty amazed that having the system come up and reconfigure appears to 
have cleared up a number of errors.  I had figured the scsi disks would 
just not be able to be accessed by the OS until I did a reconfigure.  I had 
no clue not reconfiguring would cause a ton of errors.

On a related note, I noticed something about the D1000 that may also have 
contributed to the problem.  I had attached the D1000 cable to the "second" 
SCSI controller on the JBOD, the one down by the last numbered drive.  I 
changed this so the cable was attached next to the side with disk "0", and 
put the terminator on the other end.  Maybe this had something to do with it?

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


From slaufer at slis.sjsu.edu  Mon Jan 31 20:32:44 2005
From: slaufer at slis.sjsu.edu (Stanley Laufer)
Date: Mon, 31 Jan 2005 17:32:44 -0800 (PST)
Subject: SUMMARY: Solaris 9 /etc/shadow file, MD5 problem.
Message-ID: <Pine.GSO.4.61.0501311721020.585@slis.sjsu.edu>

Hi All,

Thank you, thank you to all who replied, since a number
of the suggestions led me in the right direction.

The consensus was that the $ indicated that the password
was encrypted using a newer encryption algorithm (most likely
MD5).

Although interestingly, the problematic /etc/shadow passwords
only had a "$1" at the beginning, instead of "$1$".

"$1$" appears to be the more common way to denote that the
entry is in MD5 encryption.  So, I'm not sure what that's
all about.

In any case, I eventually had the thought that perhaps a
recent patch had "broken" MD5.  After all, some of these MD5
passwords have been around for many months, and we have never
before had a problem.

So, on that hunch, I tracked down patch 112874-30 and uninstalled
it, and lo and behold, the MD5 password problem went away.

Several of you also indicated that I could force Sol 9 to
only use the standard UNIX DES encryption by removing MD5,
Blowfish, and Sun MD5 from CRYPT_ALGORITHMS_ALLOW in policy.conf.

I'll keep that on the back burner in case we decide to go
one way or the other (completely back to DES or completely
forward to MD5 or Blowfish).

The bottom line is that I should have done my homework when
Solaris 9 came out.  I did not realize that support for
additional encryption algorithms had been added.

Thanks again for all of your responses.  This forum has
been quite helpful over the years, and I'm sure it will
remain so for years to come.


Stanley E. Laufer
Network Administrator
School of Library and Information Science
San Jose State University
_______________________________________________
sunmanagers mailing list
sunmanagers at sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers