J2EE engine shuts down and restarts....

Question

Hi,
I am doing a scenrio using BPM with multiple parallel process. I am using a inbound file adapter for the incoming message to feed the BPM process. I have set 500 messages/per recordset.The process is stable when the message size is small but when the message size increases may be to 2MB,the J2EE engine collapses and restarts a certain period of time.
When I look at the incoming messages they seem to have failed at the BPM process. I am not sure what is causing the J2EE engine to collapse and restart...It works fine with less than 1MB incoming messages....Does it mean XI cannot support messages greater than 1MB? I know its absurd but this is very interesting..Checking the SMICM, I find the processes are all normal...
-Teresa

Answer 1

Hi :
Did u checked  up the  java heap parameters in the configtool set for the server
Raise the heap parameters as per the following steps
1) Open up the configtool and increase the heap size and memory size for the server
Cluster Data -> instance_IDXXX ->  ServerIDXXXX
Max heap Size 2048MB
Java Parameters
-Djava.security.policy=./java.policy
-Djava.security.egd=file:/dev/urandom
-Dorg.omg.CORBA.ORBClass=com.sap.engine.system.ORBProxy
-Dorg.omg.CORBA.ORBSingletonClass=com.sap.engine.system.ORBSingletonProxy
-Djavax.rmi.CORBA.PortableRemoteObjectClass=com.sap.engine.system.PortableRemoteObjectProxy
-Xdebug
-Xms1280m
-Xmx3072m
-XX:MaxPermSize=192m
-XX:NewSize=160m
-XX:MaxNewSize=160m
-XX:NewRatio=3
-XX:+DisableExplicitGC
-verbose:gc
-XX:+PrintCompilation
-XX:SurvivorRatio=2
-XX:TargetSurvivorRatio=90
-XX:PermSize=192m
-XX:+UseParNewGC
-XX:+PrintGCTimeStamps
-XX:SoftRefLRUPolicyMSPerMB=1
-Djava.awt.headless=true
2) Restart the server using Visual Administrator
3) open up the configtool again to ensure that the settings you have set in the config tool are available
4) Now check whether the problem occurs again or not
regards & cheers
karthik