Details
-
Bug
-
Status: Closed
-
Major
-
Resolution: Fixed
-
1.18.0, 1.17.1
Description
When we enable speculative execution and configure job with the following configuration:
execution.batch.speculative.enabled: true
slow-task-detector.execution-time.baseline-ratio: 0.0
slow-task-detector.execution-time.baseline-lower-bound: 0s
The ExecutionTimeBasedSlowTaskDetector will identify ExecutionJobVertex that has not yet been scheduled as slow tasks and notify them to the SpeculativeScheduler. However, the SpeculativeScheduler requires that the corresponding ExecutionVertex has entered the scheduled state before scheduling backup tasks. If this requirement is not met, it will result in speculative execution failure.
The exception stack trace is as follows:
java.lang.IllegalStateException: Execution vertex b3f44e8b1dc132ff2a47f7955c75ef7d_0 does not have a recorded version at org.apache.flink.util.Preconditions.checkState(Preconditions.java:215) ~[flink-dist-1.18-SNAPSHOT.jar:1.18-SNAPSHOT] at org.apache.flink.runtime.scheduler.ExecutionVertexVersioner.getCurrentVersion(ExecutionVertexVersioner.java:71) ~[flink-dist-1.18-SNAPSHOT.jar:1.18-SNAPSHOT] at org.apache.flink.runtime.scheduler.ExecutionVertexVersioner.lambda$getExecutionVertexVersions$1(ExecutionVertexVersioner.java:89) ~[flink-dist-1.18-SNAPSHOT.jar:1.18-SNAPSHOT] at java.util.stream.ReferencePipeline$3$1.accept(ReferencePipeline.java:193) ~[?:1.8.0_333] at java.util.HashMap$KeySpliterator.forEachRemaining(HashMap.java:1580) ~[?:1.8.0_333] at java.util.stream.AbstractPipeline.copyInto(AbstractPipeline.java:482) ~[?:1.8.0_333] at java.util.stream.AbstractPipeline.wrapAndCopyInto(AbstractPipeline.java:472) ~[?:1.8.0_333] at java.util.stream.ReduceOps$ReduceOp.evaluateSequential(ReduceOps.java:708) ~[?:1.8.0_333] at java.util.stream.AbstractPipeline.evaluate(AbstractPipeline.java:234) ~[?:1.8.0_333] at java.util.stream.ReferencePipeline.collect(ReferencePipeline.java:499) ~[?:1.8.0_333] at org.apache.flink.runtime.scheduler.ExecutionVertexVersioner.getExecutionVertexVersions(ExecutionVertexVersioner.java:90) ~[flink-dist-1.18-SNAPSHOT.jar:1.18-SNAPSHOT] at org.apache.flink.runtime.scheduler.adaptivebatch.SpeculativeScheduler.notifySlowTasks(SpeculativeScheduler.java:377) ~[flink-dist-1.18-SNAPSHOT.jar:1.18-SNAPSHOT] at org.apache.flink.runtime.scheduler.slowtaskdetector.ExecutionTimeBasedSlowTaskDetector.lambda$scheduleTask$1(ExecutionTimeBasedSlowTaskDetector.java:129) ~[flink-dist-1.18-SNAPSHOT.jar:1.18-SNAPSHOT] at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511) ~[?:1.8.0_333] at java.util.concurrent.FutureTask.run(FutureTask.java:266) ~[?:1.8.0_333] at org.apache.flink.runtime.rpc.pekko.PekkoRpcActor.lambda$handleRunAsync$4(PekkoRpcActor.java:451) ~[flink-rpc-akka48a43f0a-d73c-494a-a57b-ded9f5d82a84.jar:1.18-SNAPSHOT] at org.apache.flink.runtime.concurrent.ClassLoadingUtils.runWithContextClassLoader(ClassLoadingUtils.java:68) ~[flink-dist-1.18-SNAPSHOT.jar:1.18-SNAPSHOT] at org.apache.flink.runtime.rpc.pekko.PekkoRpcActor.handleRunAsync(PekkoRpcActor.java:451) ~[flink-rpc-akka48a43f0a-d73c-494a-a57b-ded9f5d82a84.jar:1.18-SNAPSHOT] at org.apache.flink.runtime.rpc.pekko.PekkoRpcActor.handleRpcMessage(PekkoRpcActor.java:218) ~[flink-rpc-akka48a43f0a-d73c-494a-a57b-ded9f5d82a84.jar:1.18-SNAPSHOT] at org.apache.flink.runtime.rpc.pekko.FencedPekkoRpcActor.handleRpcMessage(FencedPekkoRpcActor.java:85) ~[flink-rpc-akka48a43f0a-d73c-494a-a57b-ded9f5d82a84.jar:1.18-SNAPSHOT] at org.apache.flink.runtime.rpc.pekko.PekkoRpcActor.handleMessage(PekkoRpcActor.java:168) ~[flink-rpc-akka48a43f0a-d73c-494a-a57b-ded9f5d82a84.jar:1.18-SNAPSHOT] at org.apache.pekko.japi.pf.UnitCaseStatement.apply(CaseStatements.scala:33) [flink-rpc-akka48a43f0a-d73c-494a-a57b-ded9f5d82a84.jar:1.18-SNAPSHOT] at org.apache.pekko.japi.pf.UnitCaseStatement.apply(CaseStatements.scala:29) [flink-rpc-akka48a43f0a-d73c-494a-a57b-ded9f5d82a84.jar:1.18-SNAPSHOT] at scala.PartialFunction.applyOrElse(PartialFunction.scala:127) [flink-rpc-akka48a43f0a-d73c-494a-a57b-ded9f5d82a84.jar:1.18-SNAPSHOT] at scala.PartialFunction.applyOrElse$(PartialFunction.scala:126) [flink-rpc-akka48a43f0a-d73c-494a-a57b-ded9f5d82a84.jar:1.18-SNAPSHOT] at org.apache.pekko.japi.pf.UnitCaseStatement.applyOrElse(CaseStatements.scala:29) [flink-rpc-akka48a43f0a-d73c-494a-a57b-ded9f5d82a84.jar:1.18-SNAPSHOT] at scala.PartialFunction$OrElse.applyOrElse(PartialFunction.scala:175) [flink-rpc-akka48a43f0a-d73c-494a-a57b-ded9f5d82a84.jar:1.18-SNAPSHOT] at scala.PartialFunction$OrElse.applyOrElse(PartialFunction.scala:176) [flink-rpc-akka48a43f0a-d73c-494a-a57b-ded9f5d82a84.jar:1.18-SNAPSHOT] at scala.PartialFunction$OrElse.applyOrElse(PartialFunction.scala:176) [flink-rpc-akka48a43f0a-d73c-494a-a57b-ded9f5d82a84.jar:1.18-SNAPSHOT] at org.apache.pekko.actor.Actor.aroundReceive(Actor.scala:547) [flink-rpc-akka48a43f0a-d73c-494a-a57b-ded9f5d82a84.jar:1.18-SNAPSHOT] at org.apache.pekko.actor.Actor.aroundReceive$(Actor.scala:545) [flink-rpc-akka48a43f0a-d73c-494a-a57b-ded9f5d82a84.jar:1.18-SNAPSHOT] at org.apache.pekko.actor.AbstractActor.aroundReceive(AbstractActor.scala:229) [flink-rpc-akka48a43f0a-d73c-494a-a57b-ded9f5d82a84.jar:1.18-SNAPSHOT] at org.apache.pekko.actor.ActorCell.receiveMessage(ActorCell.scala:590) [flink-rpc-akka48a43f0a-d73c-494a-a57b-ded9f5d82a84.jar:1.18-SNAPSHOT] at org.apache.pekko.actor.ActorCell.invoke(ActorCell.scala:557) [flink-rpc-akka48a43f0a-d73c-494a-a57b-ded9f5d82a84.jar:1.18-SNAPSHOT] at org.apache.pekko.dispatch.Mailbox.processMailbox(Mailbox.scala:280) [flink-rpc-akka48a43f0a-d73c-494a-a57b-ded9f5d82a84.jar:1.18-SNAPSHOT] at org.apache.pekko.dispatch.Mailbox.run(Mailbox.scala:241) [flink-rpc-akka48a43f0a-d73c-494a-a57b-ded9f5d82a84.jar:1.18-SNAPSHOT] at org.apache.pekko.dispatch.Mailbox.exec(Mailbox.scala:253) [flink-rpc-akka48a43f0a-d73c-494a-a57b-ded9f5d82a84.jar:1.18-SNAPSHOT] at java.util.concurrent.ForkJoinTask.doExec(ForkJoinTask.java:289) [?:1.8.0_333] at java.util.concurrent.ForkJoinPool$WorkQueue.runTask(ForkJoinPool.java:1067) [?:1.8.0_333] at java.util.concurrent.ForkJoinPool.runWorker(ForkJoinPool.java:1703) [?:1.8.0_333] at java.util.concurrent.ForkJoinWorkerThread.run(ForkJoinWorkerThread.java:172) [?:1.8.0_333]
Attachments
Attachments
Issue Links
- blocks
-
FLINK-32788 SpeculativeScheduler do not handle allocate task errors when schedule speculative tasks may causes resource leakage.
- Closed
- links to