RE: Memory leak in GC 10.2.0.3.0?

From: Herring Dave - dherri <Dave.Herring_at_acxiom.com>
Date: Fri, 2 Jan 2009 14:42:15 -0600
Message-ID: <7ED53A68952D3B4C9540B4EFA5C76E3605A3B505_at_CWYMSX04.Corp.Acxiom.net>



Rich, we've run into something similar with OEM 10g (10.2.0.4). Our OMS server shares it with another DB and is an 8 x 32GB Redhat server. We've got 154 targets, of which 33 are dbs and 28 are servers (nearly all db servers).

At least once a month the OMS starts generating ORA-00600 [17059] errors and we end up having to bounce the OMS, repository, and agent on that server to stop the errors. No better resolution yet!

Dave


Dave Herring, DBA | A c x i o m M I C S / C S O

630-944-4762 office | 630-430-5988 wireless | 630-944-4989 fax

-----Original Message-----
From: oracle-l-bounce_at_freelists.org
[mailto:oracle-l-bounce_at_freelists.org] On Behalf Of Rich Jesse Sent: Monday, December 29, 2008 5:19 PM
To: oracle-l_at_freelists.org
Subject: Memory leak in GC 10.2.0.3.0?

Hey all,

I'm running Grid Control 10.2.0.3.0 on a P-IV desktop with OEL 4, 2GB RAM
and 2GB swap. I have 39 targets, with only 5 of those being Oracle DBs and
maybe 6 more hosts -- the rest is the listener, agent, beacon, etc overhead
that I have no interest in. I am generally the sole user of the console to
monitor 1 production DB.

After all of 140 days, I get an alert that swap on the OMS' server is
>97%

consumed. Sure enough, the Java slice running the OMS is taking up
>3.2GB

of virtual. I did an emergency bounce of the OMS, even though /proc/meminfo
showed a few hundred MB of FS cache which should be available for stealing,
and all's well for now.

This reeks of familiarity and sure enough, I posted something similar 10 months ago:
http://www.freelists.org/post/oracle-l/GC-10gR2-mem-leak-on-Oracle-EL4

Since I'm the only DBA here I'm thinking my "best" option is to <shudder>
*schedule* bounces of the OMS (probably quarterly), which peeves me to no
end.

Anyone care to share their OMS server's RAM usage (we've got all OMS pieces
on same server)? Am I just way short on RAM here? Is there any way to troubleshoot this? Or am I stuck bouncing the OMS?

Filing an SR is not an option, given one of the two I currently have open is
approaching it's anniversary -- I don't have that kind of time.

TIA! Rich

--
http://www.freelists.org/webpage/oracle-l


***************************************************************************
The information contained in this communication is confidential, is
intended only for the use of the recipient named above, and may be legally
privileged.

If the reader of this message is not the intended recipient, you are
hereby notified that any dissemination, distribution or copying of this
communication is strictly prohibited.

If you have received this communication in error, please resend this
communication to the sender and delete the original message or any copy
of it from your computer system.

Thank You.
****************************************************************************

--
http://www.freelists.org/webpage/oracle-l
Received on Fri Jan 02 2009 - 14:42:15 CST

Original text of this message