Response time spikes on RMI

[lie_marisa]

Hi all,

The application is running on WebLogic 12.2.1.3

Sometimes, the response time spike when calling RMI to our clustered application tier WebLogics as below

Method hotspots shows it's related with clustered rjvm

Is there anything we need to check to further analyze the issue?

Any advice would be appreciated.

Thank you!

[skrystosik · Accepted solution]

It looks like parking transaction while waiting for data. How it looks from other side (system where you are trying to connect via RMI)?

Sebastian

[lie_marisa]

In the WebLogic RMI server side, the activity traced in DynaTrace is only few ms.

It seems like the time is spent in the client(caller) side while creating the Stub, not the actual call

The question is, what would make this step slow? What should we check to find the root cause?

Thanks

[skrystosik]

You should talk about it with developers. Those are WebLogic classes, so downloading classes may not be ideal option here. But you can always try this option.

Good idea may be as well checking WebLogic community, because this may be issue related to particular version.

Sebastian

[lie_marisa]

I'll try WebLogic community. Thank you