Networker

Re: [Networker] Problem with Legato 6.1.3

2003-07-03 06:51:51
Subject: Re: [Networker] Problem with Legato 6.1.3
From: Stan Horwitz <stan AT TEMPLE DOT EDU>
To: NETWORKER AT LISTMAIL.TEMPLE DOT EDU
Date: Thu, 3 Jul 2003 06:51:00 -0400
On Thu, 3 Jul 2003, John Gowing wrote:

>Good Day Claudio.
>
>I don't have much to add by way of solution at this point, but can say
>that I am having the same/similar problem. I have a an IBM 3494 Silo with
>FC attach 3590 drives, also running on 6.1.3 on AIX 5.1. We experience
>the problem both with backups and with a cloning operation. The cloning
>happens completely on the NW Server, and thus eliminates the network as a
>potential cause. When the job slows or hangs we see very high utilisation
>on the associated NSRMMD task. Stopping the Job, Killing the NSRMMD
>concerned, allowing the server to restart the NSRMMD after the
>appropriate timeouts etc, and then restarting the job clears the problem
>and the job proceeds. However on one occasion this proceedure Hung
>Networker and we had to restart the whole Networker Server. I have logged
>a case with Legato on this, will keep you informed.

Just out of curiosity, are you seeing any nsrmmd errors in your daemon.log
too? I keep getting the error "nsrd: media info:  restart of nsrmmd #24 on
bootz cancelled". I am having a problem where the NSR server grinds to a
near halt nightly. Restarting all the nsr daemons results in much faster
performance, but only for a few hours. Rebooting the server does not help.
Legato's tech support has been feeding me suggestion after suggestion on
how to tune the Solaris 9 kernal and adjust the nsr configuration, but
nothing seems to help and the issue remains unresolved. I have also been
running nsrmmd in debug mode for the past week or so and supplying NSR's
tech support with the debug data daily.

I am using NSR 6.1.3 too, on Solaris 9 as the server and nmdp with
SnapImage on our NetWorker server. I wonder if there is a problem with
SnapImage and/or nmdmp. The server is a Sun E450 with a single CPU. This
is getting very frustrating because the system ran fine prior to our
migration from Tru64 Unix to Solaris 9 a month ago and simultaneously
installing SnapImage to back up a pair of new Mirapoint message stores.
The only way I have been able to avoid this problem is to run the ndmp
backups for our two Mirapoint servers during the day and not run them at
night when the bulk of our backups are done.

Legato's tech support people are still working on this issue, but it is
very frustrating having to wait while this problem is being worked out.
I am starting to wonder if it would be prudent to downgrade to an earlier
NSR version on our Legato server, or perhaps upgrade to NetWorker 7.

--
Note: To sign off this list, send a "signoff networker" command via email
to listserv AT listmail.temple DOT edu or visit the list's Web site at
http://listmail.temple.edu/archives/networker.html where you can
also view and post messages to the list.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=