[FLINK-19552] JobManager dies with IllegalStateException SharedSlot (physical request SlotRequestId{%}) has been released - ASF JIRA

Details

Type: Bug
Status: Closed
Priority: Blocker
Resolution: Fixed
Affects Version/s: 1.12.0
Fix Version/s: 1.12.0
Component/s: Runtime / Coordination
Labels:
- pull-request-available

Description

While running some benchmarks that involve a lot of failures, I experienced fatal JobManager crashes, with the following log:

2020-10-09 09:01:45,001 INFO  org.apache.flink.runtime.executiongraph.ExecutionGraph       [] - CHAIN Filter (Filter at main(TPCHQuery3.java:157)) -> Map (Map at appendMapper(KillerClientMapper.java:38)) (2/8) (cf993790aa641a2287b42939b3037f75_f25e1269f08c88185d8a3b9caad8d0c0_1_2) switched from CREATED to SCHEDULED.
2020-10-09 09:01:45,004 ERROR org.apache.flink.runtime.util.FatalExitExceptionHandler      [] - FATAL: Thread 'flink-akka.actor.default-dispatcher-17' produced an uncaught exception. Stopping the process...
java.util.concurrent.CompletionException: java.lang.IllegalStateException: SharedSlot (physical request SlotRequestId{214e8e202b9b087388a6a17c6ba9bccf}) has been released
        at java.util.concurrent.CompletableFuture.encodeThrowable(CompletableFuture.java:273) ~[?:1.8.0_222]
        at java.util.concurrent.CompletableFuture.completeThrowable(CompletableFuture.java:280) ~[?:1.8.0_222]
        at java.util.concurrent.CompletableFuture.uniRun(CompletableFuture.java:708) ~[?:1.8.0_222]
        at java.util.concurrent.CompletableFuture$UniRun.tryFire(CompletableFuture.java:687) ~[?:1.8.0_222]
        at java.util.concurrent.CompletableFuture$Completion.run(CompletableFuture.java:442) ~[?:1.8.0_222]
        at org.apache.flink.runtime.rpc.akka.AkkaRpcActor.handleRunAsync(AkkaRpcActor.java:402) ~[flip1-bench-jobs-1.0-SNAPSHOT.jar:?]
        at org.apache.flink.runtime.rpc.akka.AkkaRpcActor.handleRpcMessage(AkkaRpcActor.java:195) ~[flip1-bench-jobs-1.0-SNAPSHOT.jar:?]
        at org.apache.flink.runtime.rpc.akka.FencedAkkaRpcActor.handleRpcMessage(FencedAkkaRpcActor.java:74) ~[flip1-bench-jobs-1.0-SNAPSHOT.jar:?]
        at org.apache.flink.runtime.rpc.akka.AkkaRpcActor.handleMessage(AkkaRpcActor.java:152) ~[flip1-bench-jobs-1.0-SNAPSHOT.jar:?]
        at akka.japi.pf.UnitCaseStatement.apply(CaseStatements.scala:26) [flip1-bench-jobs-1.0-SNAPSHOT.jar:?]
        at akka.japi.pf.UnitCaseStatement.apply(CaseStatements.scala:21) [flip1-bench-jobs-1.0-SNAPSHOT.jar:?]
        at scala.PartialFunction$class.applyOrElse(PartialFunction.scala:123) [flip1-bench-jobs-1.0-SNAPSHOT.jar:?]
        at akka.japi.pf.UnitCaseStatement.applyOrElse(CaseStatements.scala:21) [flip1-bench-jobs-1.0-SNAPSHOT.jar:?]
        at scala.PartialFunction$OrElse.applyOrElse(PartialFunction.scala:170) [flip1-bench-jobs-1.0-SNAPSHOT.jar:?]
        at scala.PartialFunction$OrElse.applyOrElse(PartialFunction.scala:171) [flip1-bench-jobs-1.0-SNAPSHOT.jar:?]
        at scala.PartialFunction$OrElse.applyOrElse(PartialFunction.scala:171) [flip1-bench-jobs-1.0-SNAPSHOT.jar:?]
        at akka.actor.Actor$class.aroundReceive(Actor.scala:517) [flip1-bench-jobs-1.0-SNAPSHOT.jar:?]
        at akka.actor.AbstractActor.aroundReceive(AbstractActor.scala:225) [flip1-bench-jobs-1.0-SNAPSHOT.jar:?]
        at akka.actor.ActorCell.receiveMessage(ActorCell.scala:592) [flip1-bench-jobs-1.0-SNAPSHOT.jar:?]
        at akka.actor.ActorCell.invoke(ActorCell.scala:561) [flip1-bench-jobs-1.0-SNAPSHOT.jar:?]
        at akka.dispatch.Mailbox.processMailbox(Mailbox.scala:258) [flip1-bench-jobs-1.0-SNAPSHOT.jar:?]
        at akka.dispatch.Mailbox.run(Mailbox.scala:225) [flip1-bench-jobs-1.0-SNAPSHOT.jar:?]
        at akka.dispatch.Mailbox.exec(Mailbox.scala:235) [flip1-bench-jobs-1.0-SNAPSHOT.jar:?]
        at akka.dispatch.forkjoin.ForkJoinTask.doExec(ForkJoinTask.java:260) [flip1-bench-jobs-1.0-SNAPSHOT.jar:?]
        at akka.dispatch.forkjoin.ForkJoinPool$WorkQueue.runTask(ForkJoinPool.java:1339) [flip1-bench-jobs-1.0-SNAPSHOT.jar:?]
        at akka.dispatch.forkjoin.ForkJoinPool.runWorker(ForkJoinPool.java:1979) [flip1-bench-jobs-1.0-SNAPSHOT.jar:?]
        at akka.dispatch.forkjoin.ForkJoinWorkerThread.run(ForkJoinWorkerThread.java:107) [flip1-bench-jobs-1.0-SNAPSHOT.jar:?]
Caused by: java.lang.IllegalStateException: SharedSlot (physical request SlotRequestId{214e8e202b9b087388a6a17c6ba9bccf}) has been released
        at org.apache.flink.util.Preconditions.checkState(Preconditions.java:217) ~[flip1-bench-jobs-1.0-SNAPSHOT.jar:?]
        at org.apache.flink.runtime.scheduler.SharedSlot.allocateLogicalSlot(SharedSlot.java:126) ~[flip1-bench-jobs-1.0-SNAPSHOT.jar:?]
        at org.apache.flink.runtime.scheduler.SlotSharingExecutionSlotAllocator.allocateLogicalSlotsFromSharedSlots(SlotSharingExecutionSlotAllocator.java:165) ~[flip1-bench-jobs-1.0-SNAPSHOT.jar:?]
        at org.apache.flink.runtime.scheduler.SlotSharingExecutionSlotAllocator.allocateSlotsFor(SlotSharingExecutionSlotAllocator.java:127) ~[flip1-bench-jobs-1.0-SNAPSHOT.jar:?]
        at org.apache.flink.runtime.scheduler.DefaultScheduler.allocateSlots(DefaultScheduler.java:364) ~[flip1-bench-jobs-1.0-SNAPSHOT.jar:?]
        at org.apache.flink.runtime.scheduler.DefaultScheduler.allocateSlotsAndDeploy(DefaultScheduler.java:337) ~[flip1-bench-jobs-1.0-SNAPSHOT.jar:?]
        at org.apache.flink.runtime.scheduler.strategy.PipelinedRegionSchedulingStrategy.maybeScheduleRegion(PipelinedRegionSchedulingStrategy.java:143) ~[flip1-bench-jobs-1.0-SNAPSHOT.jar:?]
        at org.apache.flink.runtime.scheduler.strategy.PipelinedRegionSchedulingStrategy.maybeScheduleRegions(PipelinedRegionSchedulingStrategy.java:128) ~[flip1-bench-jobs-1.0-SNAPSHOT.jar:?]
        at org.apache.flink.runtime.scheduler.strategy.PipelinedRegionSchedulingStrategy.restartTasks(PipelinedRegionSchedulingStrategy.java:100) ~[flip1-bench-jobs-1.0-SNAPSHOT.jar:?]
        at org.apache.flink.runtime.scheduler.DefaultScheduler.lambda$restartTasks$2(DefaultScheduler.java:290) ~[flip1-bench-jobs-1.0-SNAPSHOT.jar:?]
        at java.util.concurrent.CompletableFuture.uniRun(CompletableFuture.java:705) ~[?:1.8.0_222]
        ... 24 more