Re: [nv-l] ovwdb high CPU after upgrade to 7.1.4
2005-01-26 08:37:38
Ooops! My apologies. It's been pointed out to me that I'm blind and that your entry is fifty thousand not five thousand.
Of course, despite the fact that you just migrated to 7.1.4 you are still a full 2 fixpacks behind on maintenance and the third will be out at the end of March. You plain old 7.1. has been out of support for a couple of years now.
Don't know what else to tell you as ovwdb is not my area, but there are some things you could look at.
(1) Have you done ovtopofix lately? That's always a good idea when the GUI is synchronizing too frequently.
(2) smartsets. Are you using them for community name resolution in xnmsnmpconf? That's OK but after you migrate the cache for it is empty and has to be rebuilt. So you would have netmon bugging nvcold and him bugging ovwdb for answers, If you started the GUI immediately after the upgrade, then collmap would be doing the same thing. Don't know how long you let netmon run before starting there GUI after the upgrade but that could account for the high usage. Has it leveled off any?
If not, you should probably contact Support, though I am almost certain they will recommend that you apply FP02. It's exceptionally difficult to troubleshoot a problem when the system with the problem is not at the level which the troubleshooter has.
James Shanks
Level 3 Support for Tivoli NetView for UNIX and Windows
Tivoli Software / IBM Software Group
James Shanks/Raleigh/IBM@IBMUS
James Shanks/Raleigh/IBM@IBMUS
Sent by: owner-nv-l AT lists.us.ibm DOT com
01/25/2005 08:38 PM
|
|
Well, you may have checked it, but it's wrong
the entry you show says " -n5000"
for 25000 objects it should say "-n30000" which is 25000 plus 20 per cent
James Shanks
Level 3 Support for Tivoli NetView for UNIX and Windows
Tivoli Software / IBM Software Group
"Jorge A. Jiles" <jorge.jiles AT ualberta DOT ca>
"Jorge A. Jiles" <jorge.jiles AT ualberta DOT ca>
Sent by: owner-nv-l AT lists.us.ibm DOT com
01/25/05 06:02 PM
|
|
Yes, I checked that already. Bellow is the configuration of ovwdb.lrf. I'm not familiar with what is in the FFDC directory.
OVs_YES_START:nvsecd:-O,-n50000,-t:OVs_WELL_BEHAVED:15:/usr/OV/conf/FFDC/scripts/ovwdb_FFDC:5:
At 03:44 PM 1/25/2005, you wrote:
Have you checked the ovwdb cache size to make sure it is greater than the 25000 objects? The default is 5000. Change it with serversetup.
James Shanks
Level 3 Support for Tivoli NetView for UNIX and Windows
Tivoli Software / IBM Software Group
"Jorge A. Jiles" <jorge.jiles AT ualberta DOT ca>
"Jorge A. Jiles" <jorge.jiles AT ualberta DOT ca> Sent by: owner-nv-l AT lists.us.ibm DOT com
01/25/2005 05:33 PM Please respond to nv-l
To
nv-l AT lists.us.ibm DOT com
cc
Subject
[nv-l] ovwdb high CPU after upgrade to 7.1.4
Netview 7.1.4
Solaris 8
Sun blade 1000 2x750 CPU
2 GB RAM
Approximately 25000 objects in the database
I just upgraded Netview from 7.1 to 7.1.4. The upgrade seemed ok, no errors
during the installations or the log files. The problem I'm having is that
now ovwdb process uses as much as 50% of CPU processing. The GUI
synchronizes constantly making the application very slow in response time.
I did the same upgrade in my test system with no problems, although it runs
a much smaller database.
I think I read somewhere a while ago about this situation, but I can not
remember what the recommended solution was. I can not find this problem in
the release notes.
If any one has any recommendation in how to solve this problem, it will be
highly appreciated.
Thanks
.
Jorge A Jiles
Network Analyst
Computing & Network Services
University of Alberta
Edmonton, Alberta
Canada
Jorge A Jiles
Network Analyst
Computing & Network Services
University of Alberta
Edmonton, Alberta
Canada
|
|
|