ADSM-L

Re: Long pause in the middle of an NT backup.

2004-06-08 10:23:23
Subject: Re: Long pause in the middle of an NT backup.
From: Ben Bullock <bbullock AT MICRON DOT COM>
To: ADSM-L AT VM.MARIST DOT EDU
Date: Tue, 8 Jun 2004 08:22:58 -0600
        If the pause happens between disk drives, typically this is the
period that the TSM server is sending a list of the "active" files that
it saw on the previous backup. If it's taking a long time, it could mean
a few things:
        - Possible misconfiguration of the NIC causing it to go slow.
        - Very large list of files being sent.
        - The large list of files is being cached to swap because of
limited resources on the client.

The best way to see what is happening during that gap is to do some "q
sess" on the TSM server during the pause. What does the TSM server say
is going on?
        - Is it in a RUN state?
        - Is the TSM server sending a large amount of data to the
client?
        - Is it in an "SendW" or "IdleW" state? That may point to a NIC
or resource issue on the client.

Ben


-----Original Message-----
From: ADSM: Dist Stor Manager [mailto:ADSM-L AT VM.MARIST DOT EDU] On Behalf Of
Anderson, Matt
Sent: Tuesday, June 08, 2004 7:46 AM
To: ADSM-L AT VM.MARIST DOT EDU
Subject: Long pause in the middle of an NT backup.


I'm experiencing behavior on some NT nodes where, in the middle of a
scheduled backup, there will be a long pause where no activity seems to
occur. It would appear to occur after the backup of one drive finishes,
but before the next one starts. For example, see this log excerpt from
this
morning:

06/08/2004 01:30:46 ANS1898I ***** Processed 1,911,000 files *****
06/08/2004 01:30:47 ANS1898I ***** Processed 1,916,000 files *****
06/08/2004 01:30:49 ANS1898I ***** Processed 1,916,500 files *****
06/08/2004 01:31:50 Normal File-->       141,971,832
\\my-server\f$\PerforceETC\Logs\Log.txt  Changed
06/08/2004 01:32:34 Retry # 1  Normal File-->       141,971,832
\\my-server\f$\PerforceETC\Logs\Log.txt [Sent]
06/08/2004 01:32:34 Successful incremental backup of '\\my-server\f$'

06/08/2004 07:56:10 ANS1898I ***** Processed 1,917,000 files *****
06/08/2004 07:56:13 ANS1898I ***** Processed 1,917,500 files *****
06/08/2004 07:56:15 ANS1898I ***** Processed 1,918,000 files *****
06/08/2004 07:56:17 ANS1898I ***** Processed 1,918,500 files *****

There's almost 6.5 hours where no activity was logged in the
dsmsched.log. There is a (possibly) corresponding message in the
dsmerror.log:

06/08/2004 01:30:48 win32NpWrite(): Error 233 writing to named pipe

The client is Windows 2000 running TSM client version 5.2.2.5 The TSM
server is AIX 5.2 running TSM server version 5.2.2.2

Any insight that can be provided would be appreciated!

Thanks,

Matt Anderson
Backup & Recovery Engineer

************************************************************************
****
This email may contain confidential material.
If you were not an intended recipient,
Please notify the sender and delete all copies.
We may monitor email to and from our network.
************************************************************************
****