This can occur in a clustered environment where you submit create identity task and sporadically the task will fail some of the clustered nodes.
[5/7/17 15:54:44:020 EDT] 00004ced SibMessage W [:] CWSIV0776W: A connection to remote messaging engine EUA-CAIM.000-iam_im-IMSBus on bus iam_im-IMSBus for endpoint activation [com.ibm.ws.sib.ra.inbound.impl.SibRaScalableEndpointActivation@fc68b3f7 <active=true> <connections={2E0AE9F31AA4A0F1=com.ibm.ws.sib.ra.inbound.impl.SibRaMessagingEngineConnection@9baec364}> <messageEndpointFactory=com.ibm.ejs.container.MessageEndpointFactoryImpl@bc3dfcac> <endpointConfiguration=[com.ibm.ws.sib.api.jmsra.impl.JmsJcaActivationSpecImpl$JmsJcaEndpointConfigurationImpl@-889551577 <JmsJcaActivationSpecImpl.this=[com.ibm.ws.sib.api.jmsra.impl.JmsJcaActivationSpecImpl@437165920 <userName=null> <password=null> <xaRecoveryAlias=null> <destination=topic://iam_im-ServerCommand?busName=iam_im-IMSBus> <durableSubscriptionHome=EUA-CAIM.000-iam_im-IMSBus> <destinationType=javax.jms.Topic> <messageSelector=MessageFormat = '1.0'> <acknowledgeMode=Auto-acknowledge> <subscriptionDurability=NonDurable>
javax.transaction.RollbackException.
Server.log returns:
[5/7/17 19:08:33:903 EDT] 00000245 ServletWrappe E com.ibm.ws.webcontainer.servlet.ServletWrapper service SRVE0068E: An exception was thrown by one of the service methods of the servlet [/app/cui7/home.jsp] in application [iam_im]. Exception created : [java.lang.NullPointerException
at java.text.SimpleDateFormat.<init>(SimpleDateFormat.java:625)
at java.text.DateFormat.get(DateFormat.java:784)
This issue was be caused by a discrepancy in the time between the servers. This can also be caused by a custom PXpolicy or JavaScript.
Check all clustered nodes. Is the date\time synced? Are they exactly the same and synchronized through NTP?
https://en.wikipedia.org/wiki/Network_Time_Protocol
The nodes not matching the date\time exactly caused the issue. Even a slight deviation can cause sporadic values depending on the date\time needed being used in the task. This could cause the time functions to report as null as we were seeing here but also not always happen depending on the criteria of the task. A PXpolicy or Javascript could return different values depending on the nodes because of time not synced.
Drill into a failing task and look at event. Review which event failed and depending on the failed review the change values of the event. Do you have a custom PXpolicy or Javascript date\time causing the issue?
Restart the problematic nodes. Monitor the nodes that were effected restarted for any of the failed create user tasks. If another factor caused a node sync issue restarting would also help resolve the issue. This would also resolve any PXpolicy issues that could return NULL values due to a clock sync issue.