SMDiagnostics - setup

[Former Member]

I have been setting up a Solution Manager Diagnostics system. The hotpacks are all at the right levels and the java SP's are all correct. I have completed the initial configuration of the SMD and have configured a solution with many of my systems in it.

In the SMD I go to Diagnostics Setup then Managed Systems and click on Setup Wizard. My solution shows up with all of the correct systems in it. I click on the system I want to configure and confirm that all the JAVA connection parameters are correct then click next.

SMD runs off and does all manner of configurations, but its failing on the DataCollect phase (The initial Data Collect will be run on managed systems). Everything else completes without error - green lights are everywhere. But the initial data collection keeps failing with this error:

Step DataCollect Details 
 
Initial Data Collect WARNING for host tewk-epdv.global.avidww.com
!! Exception : Thu Mar 05 15:09:49 EST 2009 | startInstance:Instance [Server][]: Attributes <Type> or <ID> cannot be null or empty
 
 Exceptions 
com.sap.sup.admin.scheduler.exception.WarnTaskExecutionException: Thu Mar 05 15:09:49 EST 2009 | startInstance:Instance [Server][]: Attributes <Type> or <ID> cannot be null or empty

It looks to me like these attributes, Type and/or ID are empty but for the life of me I can't find where it might be. I have gone through transaction SMSY with a fine-toothed comb looking for a value that can be filled in, but isn't.

Does anyoen have any idea what this might be referring to?

Thanks.

-Rob Solomon

[raguraman_c · Accepted solution]

Hi,

This note

1096323 - SMD Agent DataCollector Application - Common errors

should solve your problem.

Feel free to revert back.

-=-Ragu

[Former Member]

Hi,

Can i get some response for this if anyone knows about this issue ?

Thanks,

Murali.

[Former Member]

The following is what the detail log shown after clicking the error from scheduler:

3028] The task Collect Configuration Data (sappp1-v3p) failed during its processing!

[EXCEPTION]

com.sap.sup.admin.scheduler.exception.FatalTaskExecutionException: An exception occurred while parsing file fc.xml from host [sappp1-v3p] !

at com.sap.sup.admin.scheduler.task.remote.config.ConfigGatherTask.process_FilesystemConfiguration(ConfigGatherTask.java(Compiled Code))

at com.sap.sup.admin.scheduler.task.remote.config.ConfigGatherTask.process(ConfigGatherTask.java(Compiled Code))

at sun.reflect.GeneratedMethodAccessor232.invoke(Unknown Source)

at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java(Compiled Code))

at java.lang.reflect.Method.invoke(Method.java(Compiled Code))

at com.sap.smd.server.exec.asio.AsioInvocationHandler$1.call(AsioInvocationHandler.java(Compiled Code))

at com.sap.smd.server.util.concurrent.FutureResult$1.run(FutureResult.java(Compiled Code))

at com.sap.smd.server.exec.asio.AsioInvocationHandler$AsioRunner.run(AsioInvocationHandler.java(Compiled Code))

at com.sap.smd.server.exec.TaskRunner.run(TaskRunner.java(Compiled Code))

at com.sap.smd.server.util.concurrent.PooledExecutor$Worker.run(PooledExecutor.java(Compiled Code))

at java.lang.Thread.run(Thread.java(Compiled Code))

Caused by: java.lang.NullPointerException

Thanks,

Murali.

[Former Member]

Hi,

I could able to get rid of this warning finally. I renamed the server1 folder (not usable one) but did not move it out of the cluster folder, after doing that and restarting the agent again, the Attributes warning gone.

But still my daily Collect Configuration Data scheduler job fails :

java.lang.NullPointerException.

Could not find much in the logs, based on the note 1100405 checked everything , looks ok except based on remark 2) the db_sid_xxx.zip file is not creating.

Any idea about this ?

Thanks,

Murali.

[Former Member]

Hi Raghuram/Rob,

I am saving same kind of issue in my central instance and couple of app servers. (two app servers are configured good without any issue). But i went to /usr/sap/SID/JC00/j2ee/cluster. I had 2 server node dirs server0 and server1(created long back not using now) but we are using only one node server0 even from the config tool it shows only one. Still i renamed server1 to server1_old kind of thing, still setup is giving same warning, do i need to restart Portal for this change to take effect ?

Also in config tool i could not find anything related to switched off node (like server1) , is there any special place to look for that ?

Thanks,

Murali.

[Former Member]

that was it! thanks very much for your help.

last year I had added another instance in the cluster, but the filesystem was never removed. SMD was picking up on it and thought there was another instance. After I deleted the server1 directory everything worked perfectly.

And now: its miller time.