Central autoreaction method

[Former Member]

Hi All,

I've configured system monitoring scenario using CCMS in my environment. I've configured System A (Solution Manager system) as the Central Monitoring system and System B (ECC 6.0) as the monitored system. I've used central autoreaction to generate e-mail alerts in case anything goes wrong.

My problem is everything works fine till the point there is alert and when I check the status of the autoreaction method in the Central Monitoring system, it says "RUN Required". If I double click and start the method, the situation get back to same status the next time an alert occurs.

Why does the central autoreaction method behaves in such a manner? All the required jobs has been scheduled correctly and running fine.

Please help.

Regards,

Sameer

[Former Member]

Hi All,

The problem was with the CCMS agents of the sattelite system. We reassigned the CCMS agents to the CEN and that solved the issue.

[maximilian_bauer]

Hi,

how did you do this?

Regards

[Former Member]

Hello Sameer,

Is the job SAP_CCMS_MONI_BATCH_DP running? What is the frequency?

Please note the run time status: "Run required" means,

the alert was triggered, but the auto-reaction method does not start until the next time the CCMS method dispatcher (SAPMSSY8) runs. This can last up to 5 minutes if alerts are generated asynchronously (for example, by an active data collector).

However, the next time the CCMS method dispatcher runs, the auto-reaction method should be executed and the status of the node should be READY.

If the RUN_REQUIRED status does not disappear, it could be for a number of reasons:

- If MTE is present on the local system:

The auto-reaction method must not be defined as the central

auto-reaction method.

- If MTE is present on another monitored system and a central

auto-reaction method was defined:

Is one of the above true in your case?

An agent from the monitored system must be logged on to the central system, and this agent must be active. Please check whether these agents are active or not in Rz21.

You can also refer to SNOTE: 939616

Thanks,

Bidwan

[Former Member]

Hi Bidwan,

Job SAP_CCMS_MONI_BATCH_DP is running every 5 minutes. I'm also aware that the autoreaction method should automatically get to Ready status once the job runs, but thats not happening. Please see the status of autoreaction method:

ZECC91_JOBDELAY [Central Autoreaction] CEN_CHECKED Run required ( 21.07.2008 , 10:14:28

ZECC91_JOBABORT [Central Autoreaction] CEN_CHECKED Ready ( 00.00.0000 , 00:00:00 )

ZECC91_JOBDELAY [Central Autoreaction] CEN_CHECKED Ready ( 20.07.2008 , 03:29:37 )

ZECC91_JOBABORT [Central Autoreaction] CEN_CHECKED Ready ( 00.00.0000 , 00:00:00 )

ZECC91_JOBDELAY [Central Autoreaction] CEN_CHECKED Ready ( 20.07.2008 , 03:29:37 )

ZECC91_JOBABORT [Central Autoreaction] CEN_CHECKED Ready ( 00.00.0000 , 00:00:00 )

ZECC91_JOBDELAY [Central Autoreaction] CEN_CHECKED Ready ( 20.07.2008 , 03:29:37 )

ZECC91_JOBABORT [Central Autoreaction] CEN_CHECKED Ready ( 00.00.0000 , 00:00:00 )

ZECC91_JOBDELAY [Central Autoreaction] CEN_CHECKED Ready ( 00.00.0000 , 00:00:00 )

ZECC91_JOBABORT [Central Autoreaction] CEN_CHECKED Ready ( 00.00.0000 , 00:00:00 )

ZECC91_JOBDELAY [Central Autoreaction] CEN_CHECKED Run required ( 21.07.2008 , 10:09:31

ZECC91_JOBABORT [Central Autoreaction] CEN_CHECKED Ready ( 00.00.0000 , 00:00:00 )

ZECC91_JOBDELAY [Central Autoreaction] CEN_CHECKED Ready ( 00.00.0000 , 00:00:00 )

ZECC91_DUMPS [Central Autoreaction] CEN_CHECKED Run required ( 21.07.2008 , 11:27:20

As you can see the same method ZECC91_JOBDELAY has Ready status for some jobs and Run Required for some. How is that possible? Moreover you can also see that the methods are in Run Required status for quite a long time.

MTEs are in monitored system and Central Autoreaction method are defined in Central Monitoring System.

Agents are working fine. I tested them in CEN RZ21.

Any ideas?

[Former Member]

Sameer,

Job SAP_CCMS_MONI_BATCH_DP is scheduled every 5 minutes.

This is not recommended, it should be scheduled hourly (please see note 16083) or max every 30 mins not more frequently than that.

Sometimes you can see it running for a long time is that it starts each method sequentially and waits for the method to end before it will start the next one. SO, if there are too many methods that are defined to start in the background and if each of this method takes long due to DB access issues or network delays, then you will see

long execution times for this job.

If the job is scheduled too often, it can start when the previous execution is still running.

So first schedule this job every 30 mins and check..If still you have probs then schedule it every 1 hour and check. I think it will solve the problem.

Secondly,

If this doesnot solve, try defining a remote MTE in the CEN and check. Just a hit and trial check.