Apart from your answers to your questions .
Can you please check below things are tuned properly according you Environment Size(Small,Medium,Large)
-heap size all components like jserver, rate, analytical engine,dbsrv.
-Check BPPM server performance as well.
(We were facing issue partially like this i.e. Unable to load all events in console and were not able to view events properly.)
It Would be great if you log case with BMC Support to answer all queries and to get issue resolved as it seems major issue.
I did open a case on monday, today i got a suggestion; turn debug on, try to operate on an event affected by the issue and turn debug off...
I have used a hidden JSP in bppm (lesson learnt from another case) that given a event id will show if it is in the sybase db, in the cache and in the cell. The event is missing in the sybase db.
Sizing has already been checked for large env, should be ok.
Inviato dal mio dispositivo Huawei
Hello, i can't really say..we were first advised to set this variable to: bmc.proactivenet.events.serveFromJserver=false
I was concerned about this as the docs say no to use this in a large environment though.
We also had also a lot of OOB error messages in the logs and wrong monitor hierarchies (for agentless monitoring)..it seemed that these 3 behaviors (oob errors, wrong hierarchies and mismatch between jcache and cell) were somehow related.
We were suggested to run a script supplied by bmc support to resync all the patrol agents (it turned out to be the same thing of running Sync Instance Information on CMA agent per agent). We did that on those PAs that seemed to be the source of the problem (and for which many alarms showed in console as opened while they were actually closed) and in the end the issue did not happen again. It has to be said that we also applied a bunch of hotfixes since then for specific problems, but i think they had a positive side effect on the overall platform too.
There are some changes in the way we cache as the product moved into later versions.
If the system is tuned to best practice, server and database, and the system still has memory for O/S and you are still having issues then I'd log a support call
If you need to monitor the Java on TSIM then I'd implement the following article@
How to use the Generic JVM Server KM to monitor various TSIM processes - Article Number 000128133
- Sellf-Health monitoring TSOM – BMC