Interface & PO Monitoring - SoapFaultCode:4 com/sap/smd/log/E2E_MAI_Logger

[ricardocarrega_filho]

Hello folks,

I'm getting an error when attempting to "Activate Monitoring and Alerting" on Step 5.8 Complete Setup on the Interface & PO Monitoring in Solution Manager 7.1 SP10.

The log reports error "XML distribution for agent sapslpdb returned errors, see transaction SLG1: E2E_ALERTING/DCC" for some of the Data Provider Connector entries and it varies everytime I re-try this activity. Sometimes the activation works for all agents but let's say I re-try it for the same instance I just got success, it will most likely fail again what makes me think there's a problem there.

Here's what I see when I dig into t-code SLG1 --> E2E_ALERTING/DCC:

Problem class Important: Config-XML for Introscope EM : SoapFaultCode:4  com/sap/smd/log/E2E_MAI_Logger

I wonder if anyone has gone through this issue as I have searched but could not find any SCN post with such problem.

Best regards,

Ricardo

[roland_hennessy · Accepted solution]

Hi Ricardo,

Maybe SAP note 1854525 can help here?

Kidn regards,

Roland

[ricardocarrega_filho · Accepted solution]

Hi Roland, excellent suggestion. I've tried going through this note and it did not seem to work.

I then updated LMSERVICE to the latest patch, re-executed the java upgrader and re-executed the steps mentioned on SAP note 1854525 once again but the issue still persists. I know if I try it multiple times it always ends up working but what bothers me is if this consistent error on the xml distribution to the agents happens during the activation of the monitoring, in my mind it could affect the overall monitoring functionality as well... Just guessing.

Another thing I've noticed is the error below is happening to most of the agents on my monitored managed systems. Do you think this could have any relation to the previously mentioned issue?

INTERNAL ERROR: An unexpected error occured. ExtendedPooledExecutor is blocked : maximum pool s ize : 30, current pool size :61

Regards,

Ricardo

[roland_hennessy · Accepted solution]

Hi Ricardo, I guess the issue INTERNAL ERROR: An unexpected error occured. ExtendedPooledExecutor is blocked : maximum pool s ize : 30, current pool size :61 is causing the issue here. The cause is that the dispatcher thread settings is not sized properly regarding the number of agents in your landscape. You can change this value in NetWeaver Admin > Configuration > System properties. Select the dispatcher node. Update property ThreadManager >MaxThreadCount to 200. A restart of the dispatcher is needed afterwards. You can refer to the SAp Wiki Diagnostic Agent troubleshooting guide, section Scalability / Manage java stack settings for more details http://wiki.scn.sap.com/wiki/display/SMSETUP/Diagnostics+Agent+Troubleshooting?original_fqdn=wiki.sd... Kind regards, Roland

[ricardocarrega_filho · Accepted solution]

Hi Roland,

Your last response helped me resolve the ExtendedPooledExecutor issue. These errors are gone, however the original error which I reported initially on the thread still happens.

I appreciate your replies as they were very helpful on trying to resolve this issue.

Another thing I'm thinking may be related to this and a few other issues is the Introscope version which we have here being a little obsolete (9.1.0.0). I'm looking into updating it and hopefully that'll fix some things as well.

Regards,

Ricardo

[roland_hennessy · Accepted solution]

Hi Ricardo,

Yes I guess the next steps at this stage woud be to patch the Wily Introscope if it is not at the latest level.

Let me know how this goes.

Kidn regards,

Roland

[Former Member]

Hi Ricardo,

I am facing same error as you are getting while configuring PI Monitoring. I believe my IS Agent Patch is causing trouble and i am upgrading the same as specified in the note  1819577 - Workload Analysis data missing for Java systems when using ISAGENT 8.2.4.0

Will come to know if that resolves the issue tomorrow or day after post managed system restart.

Did you get any solution for the same?

Regards
Aditya

[ricardocarrega_filho]

Hi guys,

Just checking in. Does anyone have any ideas? Anyone has had this issue before?

Thanks,

Ricardo