[OOZIE-1989] NPE during a rerun with forks

• Type: Bug
• Status: Closed
• Priority: Blocker
• Resolution: Fixed
• Affects Version/s: 4.1.0
• Fix Version/s: 4.1.0
• Component/s: core
• Assignee: Robert Kanter
• Reporter: Robert Kanter
• Created: 03/Sep/14 00:23
• Updated: 12/Jan/15 06:08
• Resolved: 05/Sep/14 23:36

Description

OOZIE-1879 fixes a problem where the order of actions that the actions in a fork ended can be different than when they are executed during a rerun, resulting in an error. It does this by using a comparator to sort them into the proper order in LiteWorkflowInstance. However, the code assumes that all actions in the executionPath have an end time, which is not always true. If that happens, you get an NPE like this:

java.lang.NullPointerException
        at org.apache.oozie.workflow.lite.LiteWorkflowInstance$ActionEndTimesComparator.compare(LiteWorkflowInstance.java:739)
        at org.apache.oozie.workflow.lite.LiteWorkflowInstance$ActionEndTimesComparator.compare(LiteWorkflowInstance.java:719)
        at java.util.TimSort.countRunAndMakeAscending(TimSort.java:324)
        at java.util.TimSort.sort(TimSort.java:189)
        at java.util.TimSort.sort(TimSort.java:173)
        at java.util.Arrays.sort(Arrays.java:659)
        at java.util.Collections.sort(Collections.java:217)
        at org.apache.oozie.workflow.lite.LiteWorkflowInstance.signal(LiteWorkflowInstance.java:316)
        at org.apache.oozie.command.wf.SignalXCommand.execute(SignalXCommand.java:190)
        at org.apache.oozie.command.wf.SignalXCommand.execute(SignalXCommand.java:73)
        at org.apache.oozie.command.XCommand.call(XCommand.java:283)
        at org.apache.oozie.service.CallableQueueService$CompositeCallable.call(CallableQueueService.java:323)
        at org.apache.oozie.service.CallableQueueService$CompositeCallable.call(CallableQueueService.java:252)
        at org.apache.oozie.service.CallableQueueService$CallableWrapper.run(CallableQueueService.java:174)
        at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
        at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
        at java.lang.Thread.run(Thread.java:745)

Issue Links

• is broken by: OOZIE-1879 Workflow Rerun causes error depending on the order of forked nodes
• is related to: OOZIE-1993 Rerun fails during join in certain condition