Hugh,
We just dealt with the same exact issue. At first, we thought it was
not having ulimits set (see the 5.1 release notes), but it turned out
to be paging space that was too small when our topology DB grew. BTW,
non-well-behaved is actually unrelated to what it sounds like. I don't
know specifics, but it's OK.
Gotta be resource related...try checking swap first, then looking at
ulimits.
Best luck,
DVO
______________________________ Reply Separator _________________________________
Subject: [NV-L] Daemon errors
Author: Hugh.C.Laughlin@Cummins.com at Internet-USA
Date: 3/17/00 10:38 AM
Ok... I am having some problems here with a couple of daemons constantly failing
on me.
Our current setup is a Client-Server environment, both running AIX 4.3; NV
version 5.1.1
The daemons which keep dying on us are: ovwdb, ovtopmd, and nvcold. These
daemons, and the application as a whole, die on me when I am trying to delete
objects in the IP map. Initially I get an error mesage/window stating: Lost
connection to database. The only option here is to select OK, after which the
NV applications closes. When I try to restart NetView, I receive an error: The
following REQUIRED daemons are not running: ovwdb ovtopmd
Additional information which might help, when NetView is running, ovstatus
indicates the following two objects are not behaving well:
object manager name: trapgend
behavior: OVs_NON_WELL_BEHAVED
state: RUNNING
PID: 19420
exit status: -
object manager name: mgragentd
behavior: OVs_NON_WELL_BEHAVED
state: RUNNING
PID: 16002
exit status: -
Everything else in the ovstatus show well_behaved.
Any suggestions or idea would be greatly appreciated.
Respectfully,
Hugh C. Laughlin
Network Systems Analyst
Lockheed Martin IBS
hugh.c.laughlin@cummins.com
_________________________________________________________________________
NV-L List information (unsubscribing, policies, posting, digest version,
searchable archives): http://www.tkg.com/nv-l
|