[OOZIE-2579] Bulk kill tests in TestBulkWorkflowXCommand might fail because of a race condition

• Type: Bug
• Status: Closed
• Priority: Minor
• Resolution: Fixed
• Fix Version/s: 4.3.0
• Assignee: Peter Bacsko
• Reporter: Peter Bacsko
• Created: 20/Jun/16 11:51
• Updated: 02/Dec/16 21:04
• Resolved: 13/Jul/16 00:26

Description

There are tests in TestBulkWorkflowXCommand which perform bulk killing.

This might fail sometimes, because the externalChildIDs is set to "00000001-dummy-oozie-wrkf-W " that causes an action to fail:

org.apache.oozie.action.ActionExecutorException: IllegalArgumentException: JobId string : 00000001-dummy-oozie-wrkf-W is not properly formed
	at org.apache.oozie.action.ActionExecutor.convertException(ActionExecutor.java:443)
	at org.apache.oozie.action.hadoop.JavaActionExecutor.kill(JavaActionExecutor.java:1614)
	at org.apache.oozie.command.wf.ActionKillXCommand.execute(ActionKillXCommand.java:146)
	at org.apache.oozie.command.wf.ActionKillXCommand.execute(ActionKillXCommand.java:1)
	at org.apache.oozie.command.XCommand.call(XCommand.java:287)
	at org.apache.oozie.service.CallableQueueService$CompositeCallable.call(CallableQueueService.java:331)
	at org.apache.oozie.service.CallableQueueService$CompositeCallable.call(CallableQueueService.java:260)
	at java.util.concurrent.FutureTask.run(FutureTask.java:266)
	at org.apache.oozie.service.CallableQueueService$CallableWrapper.run(CallableQueueService.java:178)
	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
	at java.lang.Thread.run(Thread.java:745)
Caused by: java.lang.IllegalArgumentException: JobId string : 00000001-dummy-oozie-wrkf-W is not properly formed
	at org.apache.hadoop.mapreduce.JobID.forName(JobID.java:154)
	at org.apache.hadoop.mapred.JobID.forName(JobID.java:78)
	at org.apache.oozie.action.hadoop.MapReduceActionExecutor.getRunningJob(MapReduceActionExecutor.java:342)
	at org.apache.oozie.action.hadoop.JavaActionExecutor.kill(JavaActionExecutor.java:1604)
	... 10 more

Since this code runs on a separate thread, it might randomly interfere with the main test logic, which expects the job status to be "KILLED", but sometimes the ActionKillXCommand has a chance to update it to "FAILED".

Solution: set a proper (parseable) job id:

action.setExternalChildIDs("job_201601011800_0001");

See https://hadoop.apache.org/docs/r2.6.2/api/org/apache/hadoop/mapred/JobID.html