[
https://issues.apache.org/jira/browse/FLINK-4702?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15532771#comment-15532771
]
ASF GitHub Bot commented on FLINK-4702:
---------------------------------------
Github user StephanEwen commented on the issue:
https://github.com/apache/flink/pull/2559
@tzulitai Thanks for thorough review!
I don't understand the problem why the `commitSpecificOffsetsToKafka`
method is designed to commit synchronously. The `FlinkKafkaConsumerBase` has
the pending checkpoints (I think that is what you refer to). It removes the
HashMap of "offsets to commit" from the `pendingCheckpoints` Map synchronously,
before even calling the fetcher to commit.
After that, it looks to me like it does not make a difference how that Map
"offsets to commit" is used (sync or async)...
> Kafka consumer must commit offsets asynchronously
> -------------------------------------------------
>
> Key: FLINK-4702
> URL: https://issues.apache.org/jira/browse/FLINK-4702
> Project: Flink
> Issue Type: Bug
> Components: Kafka Connector
> Affects Versions: 1.1.2
> Reporter: Stephan Ewen
> Assignee: Stephan Ewen
> Priority: Blocker
> Fix For: 1.2.0, 1.1.3
>
>
> The offset commit calls to Kafka may occasionally take very long.
> In that case, the {{notifyCheckpointComplete()}} method blocks for long and
> the KafkaConsumer cannot make progress and cannot perform checkpoints.
> Kafka 0.9+ have methods to commit asynchronously.
> We should use those and make sure no more than one commit is concurrently in
> progress, to that commit requests do not pile up.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
