[
https://issues.apache.org/jira/browse/FLINK-20672?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17251694#comment-17251694
]
Yun Tang commented on FLINK-20672:
----------------------------------
I think you're right and this might happen in extra case. I think the solution
might be catching any exception when sending these abort messages just like
current disposing/discarding checkpoint in the executor.
> CheckpointAborted RPC failure can fail JM
> -----------------------------------------
>
> Key: FLINK-20672
> URL: https://issues.apache.org/jira/browse/FLINK-20672
> Project: Flink
> Issue Type: Bug
> Components: Runtime / Checkpointing
> Affects Versions: 1.12.0, 1.11.3
> Reporter: Roman Khachatryan
> Priority: Major
>
> Introduced in FLINK-8871, aborted RPC notifications are done asynchonously:
>
> {code}
> private void sendAbortedMessages(long checkpointId, long timeStamp) {
> // send notification of aborted checkpoints asynchronously.
> executor.execute(() -> {
> // send the "abort checkpoint" messages to necessary
> vertices.
> // ..
> });
> }
> {code}
> However, the executor that eventually executes this request is created as
> follows
> {code}
> final ScheduledExecutorService futureExecutor =
> Executors.newScheduledThreadPool(
> Hardware.getNumberCPUCores(),
> new ExecutorThreadFactory("jobmanager-future"));
> {code}
> ExecutorThreadFactory uses UncaughtExceptionHandler that exits JVM on error.
> cc: [~yunta]
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
