Custom WebLogic Server (WLS) Work Manager Called "Default" Throws PeerGoneException on RMI Client when Thread Becomes Stuck
(Doc ID 1074689.1)
Last updated on SEPTEMBER 28, 2020
Applies to:Oracle WebLogic Server - Version 9.2 to 10.3
Information in this document applies to any platform.
When a WebLogic Server (WLS) custom Work Manager is called "default," we have this behavior:
- A remote java RMI client calls an EJB method (via T3) on a WLS 9.2 MP2 server
- The EJB method runs request elaboration for more than 600 seconds and its thread becomes STUCK
- The remote RMI client stops waiting for response and receives a weblogic.rjvm.PeerGoneException (nested java.io.EOFException)
- Thread on server side continues the elaboration
This indicates that there was a problem communicating with the server:
Exception in thread "main" weblogic.rjvm.PeerGoneException: nested exception is:
at examples.ejb.ejb20.basic.statelessSession.statelessSession_mt9oui_EOImpl_922_WLStub.buy(Unknown Source)
With the default work manager, the RMI client waits for response apparently without limits. This is the behavior we would like to expect also with a work manager called "default."
Note that the weblogic.rjvm.PeerGoneException is not a consequence of work manager shutdown. We have inhibited work manager shutdown configuring the shutdown trigger.
To view full details, sign in with your My Oracle Support account.
Don't have a My Oracle Support account? Click to get started!
In this Document