Anyone seen halted PAM-process in 4.3 SP02 CP01?

[Andreas Eriksson]

All of a sudden we can can see some totally halted PAM-process. They can be stuck in a queued state or running but will never move forward. Restart does not resolve the problem. PAM is installed in a cluster setup with two nodes and has been working fine for over one month until now. I cannot see anything of interest in the infrastructure or in the servers. Have you also seen this problem?

 

I have raised a supportcase but just want to also check with the community.

[Andreas Eriksson]

Update: We noticed that this can occur during high load with plenty of processes of the same that run in parallel. We will implement our own queue.

[Derek (Chun) Li]

Hi Andreas,

 

You can post the question in Process Automation Community. ca process automation community

 

Regards,

Derek

[Lenn Thompson]

Thanks, Derek. I was actually able to move this to the CA Process Automation community where the team does a great job answering questions!

[deama10]

How is the heapsize and memory usage of the servers?

[Andreas Eriksson]

Both cluster nodes has 10GB of Java heap. Around 4GB is allocated.

[deama10]

Also, one other thing, do you have any problem with scriptOutput sometimes being blank? 

[Andreas Eriksson]

I am unsure Marcel. Where can I see that information? 

[deama10]

Well basically if you use pam to powershell or ssh to a server/vcenter you'll have to use the scriptOutput of the  Runscript operator. I had a similar scenario where not also everything was halted sometimes, we also didn't get the output of the of this operator

[deama10]

But you say only some processes are stuck, no all of them?

 

If all processes are stuck, it would be easier to assume that there is something wrong with the db, probably permission. But if only some processes are, it's harder. Have you tried stopping the pam service in one of the nodes?