system center operation manager: total percentage interrupt time
Hello,
Where could I find the description of the parameters of ryles and monitors in general... is it centralized?
especially this one for now:system center operation manager: total percentage interrupt time
Thanks,
DomSystem Center Operations Manager 2007 / System Center Configuration Manager 2007 R2 Support
May 26th, 2010 7:17pm
Hi,
Configuration is centralized by the RMS. You can access rules and monitors configuration in the Authoring pane in the SCOM console.
The Total Percentage Interrupt Time is a monitor (2 actually) you can find under the windows server 2003 operating system and windows server 2008 operating system targets.
Regards.
François DufourSupervize Me
Free Windows Admin Tool Kit Click here and download it now
May 26th, 2010 7:37pm
Hi Francois,
Yes I have this already but I need a description of each parameter, settings with what they are. Which value is "best practice"...?
Thanks,
DomSystem Center Operations Manager 2007 / System Center Configuration Manager 2007 R2 Support
May 26th, 2010 7:45pm
Hi,
Tell me which parameter you need information for. I don't see any specific parameters for this one only a frequency of 120s and a threshold of 10%.
You'll find further information about the monitor in the product knowledge tab.
Regards.Supervize Me
Free Windows Admin Tool Kit Click here and download it now
May 26th, 2010 7:53pm
There's a thread about this one already.
it comes down to: - the settings are good and you shouldn't have to change them if we're talking about physical servers. (aka if you come above this 10% u really should find out why)
if the servers are vm's the interrupt % is a totally useless counter (as the cpu scales with load on vm's).Rob Korving
http://jama00.wordpress.com/
May 26th, 2010 8:11pm
Hi Francois,
Yes the threshold of 10% is the factor I need to change as it seems low but how is it calculated?
Thanks,
Dom
System Center Operations Manager 2007 / System Center Configuration Manager 2007 R2 Support
Free Windows Admin Tool Kit Click here and download it now
May 26th, 2010 8:22pm
Hi Rob,
The servers are physical servers and blades.
I need to adjust the threshold as the Application owner is receiving too many alerts and from their statitics it is normal with their application!!
I have the same alerts on another type of server:
- Cluster
- Print Server
I am getting these alerts more often now than 6 months ago... with the same machines, same user numbers,
Thanks,
Dom
System Center Operations Manager 2007 / System Center Configuration Manager 2007 R2 Support
May 26th, 2010 8:24pm
10% interrupt time basically means 100ms per second the cpu is not handling threads but interrupts and it might indicate hardware problems. The knowledge tells how you can check what is causing the interrupt. if the conclusion really is that's it's
normal behaviour for these server you can set an override to what is normal for these servers (10 to xx).
Rob Korving
http://jama00.wordpress.com/
Free Windows Admin Tool Kit Click here and download it now
May 27th, 2010 4:49pm
Hi Rob,
I reviewied the alerts it seems I have at least 20 servers above this threshold of 10% so far...
And now I called the various Applications Owners and all are saying the same thing : " it is usual"
So I'm thinking if the alerts are closed within a reasonable cycle (5-10 minutes) I will just auto-resolve them.
The servers might be too small on Memory and/or Drive !!!
Thanks,
Dom
System Center Operations Manager 2007 / System Center Configuration Manager 2007 R2 Support
May 29th, 2010 9:25pm
Hey Dom
I usually have to change two things here.
Firstly I normally change the alert severity of all perfromance monitors to be warning. It does not makesense for me for a per counter to send an availability report reeling.
I also up the limit on this monitor for servres that are constantly flapping. No some may say that is the way it should be... ok thats great, but I am on-site a customer has X number of servers that this is happening to and there is no way to increase
the CPU and for those servers in that environment they behave that way...
So I up some times to as much as 20%
You can say that its not the way to go, you can tell me why but all I can say is that
1: I don't want monitors flapping and generating noise
2: I don't want my SLD/Dist apps reporting as non available where they include the full server environment and not just a number of syn transactions.
Paul Paul Keely
Free Windows Admin Tool Kit Click here and download it now
June 1st, 2010 9:25am
One further point Dom is that I assume you know about the MP
viewer
And once you hit the override for any rule/monitor it will tell you what properties you can expose to be overridden in the first place
Paul Paul Keely
June 1st, 2010 9:29am
Hi Paul,
Thanks a lot
DOmSystem Center Operations Manager 2007 / System Center Configuration Manager 2007 R2 Support
Free Windows Admin Tool Kit Click here and download it now
June 1st, 2010 4:45pm
We get this alerts only from our VM's.... so you are saying that this counter is useless for VM's, should we just exclude all of our VM's from the rule in SCOM, so we are not reported on this at all?
June 29th, 2011 11:00am