You mention "a windows 2000 clients".... is it one client or multiple
clients?
You also mentioned asking about what changed, but it isn't clear whether
"coming up blank" means nothing changed, or nothing changed that you think
would cause this problem. The implicit question is: was the client
performance always this bad? Or was it okay before, but recently starting
having problems? If the latter, when did the problems start? You might
revisit the question of what changed around that timeframe, regardless of
the significance of the change.
Other things to look at:
- What does the client options file look like?
- Provide output from "dsmc show options". Make sure to use the same
dsm.opt file as the scheduler uses.
- Provide output from "dsmc query inclexcl" Make sure to use the same
dsm.opt file as the scheduler uses.
- While the backup is running, what does a QUERY SESSION from the TSM
admin CLI show for this node's sessions?
- What else is running on the client machine while the backup is running?
- If this is the only machine having the problem (or one of relatively few
machines having the problem), compare it to a machine that runs fine: what
are the differences between the two machines? Consider installed hardware,
installed software, activity on each machine, location on the network,
etc.
- How long does it take for the "dir" command to traverse the file
system(s)? A simple script like this might be useful:
REM *** BEGIN T.CMD ***
@echo off
:TRAVERSE_DRIVE
if {%1} == {} goto :END
echo %time% - Traversing %1
dir %1\ /w/s >nul
shift
goto :TRAVERSE_DRIVE
:END
echo %time% - Done
REM *** END T.CMD ***
Invoke the script and pass in your drive letters, i.e.:
t c: e: f:
- Are you backing up any network shares?
Regards,
Andy
Andy Raibeck
IBM Software Group
Tivoli Storage Manager Client Development
Internal Notes e-mail: Andrew Raibeck/Tucson/[EMAIL PROTECTED]
Internet e-mail: [EMAIL PROTECTED]
The only dumb question is the one that goes unasked.
The command line is your friend.
"Good enough" is the enemy of excellence.
"ADSM: Dist Stor Manager" <[EMAIL PROTECTED]> wrote on 07/12/2004
10:16:54:
> I have a windows 2000 clients (v5 sp 4 build 2195) running tsm 5.1.7
> who's incremental backup never ends. Using the timestamps next to
> message: ANS1898I , it is taking between 5 minutes and 1 hour to parse
> through 500 files. I have 100+ other servers that take about 2/100 of a
> second to pass through 500 files.
>
> Tried rebooting the client and ran performance monitor. Neither helped
> or showed any problems.
>
> Here is the bad server:
>
> 07/11/2004 14:23:08 ANS1898I ***** Processed 194,500 files *****
> 07/11/2004 14:25:17 ANS1898I ***** Processed 195,000 files *****
> 07/11/2004 14:34:18 ANS1898I ***** Processed 195,500 files *****
> 07/11/2004 15:16:52 ANS1898I ***** Processed 196,000 files *****
>
> Here is a typical server:
>
> 07/04/2004 21:45:02 ANS1898I ***** Processed 649,000 files *****
> 07/04/2004 21:45:05 ANS1898I ***** Processed 649,500 files *****
> 07/04/2004 21:45:08 ANS1898I ***** Processed 650,000 files *****
> 07/04/2004 21:45:10 ANS1898I ***** Processed 650,500 files *****
>
> Does anyone have any ideas? The usual questions have been asked (what
> changed?) but we still come up blank.
>
> Thanks,
> Ralph
