In article <[EMAIL PROTECTED]>, "Pabalate, Ed"
<[EMAIL PROTECTED]> says:
>
>Our network guy complaining and stated:
>
>"I=92m experiencing a problem with ADSM with the following symptom: =
>cpu
>usage, after less than a day, is shown to be at 100% because of the
>dsmcsvc.exe process. Further, I am unable to kill the process except =
>by
Hello Ed
I happened to run into a problem that looks just like yours this past
week. We were using the adsm client 3.1.07.9 on a windows nt 4 sp 5
system running on a dell dual cpu server. And this code has been in
use since last spring. What we saw was that the backup finished
the C drive and had started working on the D drive. After 98 to 88
thousand files it stopped updating the dsmerror.log and dsmsched.log
files. The tsm server (3.7.3) had the idle time kick in 10 hours
after the last message in the dsmsched.log.
(note: The idle timeout is set looong to stop timing out the
second tsm sessions).
Stopping the scheduler service and restarting the service did not
stop the dsmcsvc.exe file that was using up a little over 50% of
the cpu. But now there were two dsmcsvc.exe tasks.
I believe that I could reproduce the problem with a manual backup.
But I was a little rushed to fit in a restart of the system so
I did not get to check as well as I liked.
I opened a trouble (etr/pmr) report with IBM/TVIOLI and they are looking
at the problem.
I upgraded the client to tsm 4.1 and the problem is still there.
I suspect that something is wrong with the ntfs file system and the
tsm client is not handling the error correctly.
If you have support, you should open a problem report with IBM/TIVOLI
Please let us know how you make out.
-------------------------------------------------------------------------
Leonard Boyle [EMAIL PROTECTED]
SAS Institute Inc. ussas4hs@ibmmail
Room RB448 [EMAIL PROTECTED]
1 SAS Campus Drive (919) 677-8000 ext 6241
Cary NC 27513
