Uploaded image for project: 'JBeret'
  1. JBeret
  2. JBERET-231

Batch thread tx context not cleaned up properly on tx timeout in chunk-type step

    XMLWordPrintable

Details

    • Hide

      I have attached a project that can be used to reproduce the issue. To do so, follow these steps:

      • In the Wildfly standalone.xml reduce the batch thread count to 1
      • Package and deploy the war
      • Execute HTTP GET http://localhost:8080/start (this will start a batch job causing a transaction timeout after 5 seconds)
      • Execute HTTP GET http://localhost:8080/start one more time and note that the job fails when attempting to start a transaction
      Show
      I have attached a project that can be used to reproduce the issue. To do so, follow these steps: In the Wildfly standalone.xml reduce the batch thread count to 1 Package and deploy the war Execute HTTP GET http://localhost:8080/start (this will start a batch job causing a transaction timeout after 5 seconds) Execute HTTP GET http://localhost:8080/start one more time and note that the job fails when attempting to start a transaction

    Description

      When a transaction timeout occurs in a batch thread running a chunk-type step, the transaction context does not seem to be cleaned up properly. New chunk-type steps cannot be started on the respective thread any more - a server restart is required.

      Following exception is thrown when trying to run a chunk-type step in a batch thread that experienced a tx timeout before:

      12:10:48,070 INFO  [org.jberet.test.StepBoundaryLogger] (Batch Thread - 1) Starting job job
12:10:48,071 INFO  [org.jberet.test.StepBoundaryLogger] (Batch Thread - 1) Starting step jobStep
12:10:48,072 TRACE [com.arjuna.ats.jta] (Batch Thread - 1) BaseTransaction.begin
12:10:48,072 TRACE [com.arjuna.ats.jta] (Batch Thread - 1) TransactionImple.getStatus: javax.transaction.Status.STATUS_ROLLEDBACK
12:10:48,076 ERROR [org.jberet] (Batch Thread - 1) item-count=20, time-limit=0, skip-limit=-1, skipCount=0, retry-limit=-1, retryCount=0
12:10:48,076 ERROR [org.jberet] (Batch Thread - 1) JBERET000007: Failed to run job job, jobStep, org.jberet.job.model.Chunk@3a5ab142: javax.transaction.NotSupportedException: BaseTransaction.checkTransactionState - ARJUNA016051: thread is already associated with a transaction!
	at com.arjuna.ats.internal.jta.transaction.arjunacore.BaseTransaction.begin(BaseTransaction.java:73)
	at com.arjuna.ats.jbossatx.BaseTransactionManagerDelegate.begin(BaseTransactionManagerDelegate.java:78)
	at org.jberet.runtime.runner.ChunkRunner.run(ChunkRunner.java:187)
	at org.jberet.runtime.runner.StepExecutionRunner.runBatchletOrChunk(StepExecutionRunner.java:219)
	at org.jberet.runtime.runner.StepExecutionRunner.run(StepExecutionRunner.java:140)
	at org.jberet.runtime.runner.CompositeExecutionRunner.runStep(CompositeExecutionRunner.java:164)
	at org.jberet.runtime.runner.CompositeExecutionRunner.runFromHeadOrRestartPoint(CompositeExecutionRunner.java:88)
	at org.jberet.runtime.runner.JobExecutionRunner.run(JobExecutionRunner.java:59)
	at org.wildfly.extension.batch.jberet.impl.BatchEnvironmentService$WildFlyBatchEnvironment$1.run(BatchEnvironmentService.java:243)
	at org.wildfly.extension.requestcontroller.RequestController$QueuedTask$1.run(RequestController.java:497)
	at org.jberet.spi.JobExecutor$3.run(JobExecutor.java:161)
	at org.jberet.spi.JobExecutor$1.run(JobExecutor.java:99)
	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
	at java.lang.Thread.run(Thread.java:745)
	at org.jboss.threads.JBossThread.run(JBossThread.java:320)
Caused by: java.lang.IllegalStateException: BaseTransaction.checkTransactionState - ARJUNA016051: thread is already associated with a transaction!
	at com.arjuna.ats.internal.jta.transaction.arjunacore.BaseTransaction.checkTransactionState(BaseTransaction.java:264)
	at com.arjuna.ats.internal.jta.transaction.arjunacore.BaseTransaction.begin(BaseTransaction.java:68)
	... 15 more

12:10:48,077 INFO  [org.jberet.test.StepBoundaryLogger] (Batch Thread - 1) Step jobStep finished
12:10:48,077 INFO  [org.jberet.test.StepBoundaryLogger] (Batch Thread - 1) Job job finished

      Attachments

        Issue Links

          is related to

          Bug - A problem that impairs or prevents the functions of the product. JBEAP-4811 (7.0.z) Batch thread tx context not cleaned up properly on tx timeout in chunk-type step

          • Major - To be worked after higher priority work items are addressed. When the severity is high and the effort to change it is low to moderate, this term is chosen. Issues in this category likely have an existing workaround but implementation or execution may be non-trivial.
          • Verified
          links to

          Web Link After the transaction reaper has run , ChunkRunner does not clear the transaction from the thread

          Activity

            People

              cfang@redhat.com Cheng Fang
              moritz.becker@gmx.at Moritz Becker (Inactive)
              Votes:
              0 Vote for this issue
              Watchers:
              2 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved: