ovstart, ovstop, and ovstatus are commands to ovspmd. The command
interface is single-threaded, so if you do an ovstart, and then an ovstatus
before it completes, the ovstatus will hang waiting its chance to run.
This is normal.
If you think you have problems with one daemon, then try starting them one
at a time, or a few at a time, to help isolate the culprit. For a
suspicion that ovtopmd is a problem, I would do
ovstart nvsecd
ovstart ovwdb
ovstart pmd
ovstart trapd
and finally, ovstart ovtopmd
and see where the bottleneck is. If ovtopmd is really the problem, then I
recommend a general database cleanup.
I'd wait for ovtopmd to come up, however long it takes, do a plain ovstart
to bring up the rest of the daemons, and then run (1) ovmapcount -a and
(2) ovtopofix -A (-a if you have more than one map).
If that doesn't help, then please call Support.
James Shanks
Tivoli (NetView for UNIX) L3 Support
gc080@CIVILITY.AMSTERDAM.NL on 06/22/98 05:59:52 AM
Please respond to NV-L@UCSBVM.UCSB.EDU
To: NV-L@UCSBVM.UCSB.EDU
cc: (bcc: James Shanks)
Subject: Problem with ovtopmd
Hi,
Several times already I can't start NetView
when I try to run ovstatus ovstatus locks up
with ps -ef | grep '\[' I get:
root 8538 10808 63 Jun 21 - 121:13 [ovtopmd]
The only thing that helps is a reboot, but this is not the preferred
method on a production machine.
Can someone help me?
Judha
|