[
https://issues.apache.org/jira/browse/FLINK-6321?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15976404#comment-15976404
]
Stefan Richter commented on FLINK-6321:
---------------------------------------
Ok, then this problem can be related. The easiest explanation would be that the
RocksDB checkpoint got cancelled because the job failed again in the meantime.
If the checkpoint had failed for another reason than this, you should find a
younger exception in the log.
My theory is: RocksDB checkpoint takes longer than for the heap backend, so job
cancelation due to the above exception causes the RocksDB checkpoint to be
canceled. The checkpoint on the heap backend might run faster and complete
before cancelation happens. Could that make sense?
In any case, I suggest that we first fix the CNF exception and then check if
this problem still exists.
> RocksDB state backend Checkpointing is not working with KeydCEP.
> ----------------------------------------------------------------
>
> Key: FLINK-6321
> URL: https://issues.apache.org/jira/browse/FLINK-6321
> Project: Flink
> Issue Type: Bug
> Components: CEP, State Backends, Checkpointing
> Affects Versions: 1.2.0
> Environment: yarn-cluster, RocksDB State backend, Checkpointing every
> 1000 ms
> Reporter: Shashank Agarwal
>
> Checkpointing is not working with RocksDBStateBackend when using CEP. It's
> working fine with FsStateBackend and MemoryStateBackend. Application failing
> every-time.
> {code}
> 04/18/2017 21:53:20 Job execution switched to status FAILING.
> AsynchronousException{java.lang.Exception: Could not materialize checkpoint
> 46 for operator KeyedCEPPatternOperator -> Map (1/4).}
> at
> org.apache.flink.streaming.runtime.tasks.StreamTask$AsyncCheckpointRunnable.run(StreamTask.java:980)
> at
> java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
> at java.util.concurrent.FutureTask.run(FutureTask.java:266)
> at
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
> at
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
> at java.lang.Thread.run(Thread.java:745)
> Caused by: java.lang.Exception: Could not materialize checkpoint 46 for
> operator KeyedCEPPatternOperator -> Map (1/4).
> ... 6 more
> Caused by: java.util.concurrent.CancellationException
> at java.util.concurrent.FutureTask.report(FutureTask.java:121)
> at java.util.concurrent.FutureTask.get(FutureTask.java:192)
> at
> org.apache.flink.util.FutureUtil.runIfNotDoneAndGet(FutureUtil.java:40)
> at
> org.apache.flink.streaming.runtime.tasks.StreamTask$AsyncCheckpointRunnable.run(StreamTask.java:915)
> ... 5 more
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
