[
https://issues.apache.org/jira/browse/FLINK-40912?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18125278#comment-18125278
]
Sergey Nuyanzin commented on FLINK-40912:
-----------------------------------------
Merged as
[e47cf425dcd6bd8b8d8d7d65b1f740b05f00ee8e|https://github.com/apache/flink-connector-kafka/commit/e47cf425dcd6bd8b8d8d7d65b1f740b05f00ee8e]
> ExactlyOnceKafkaWriter never closes the AdminClient it creates for the
> listing abort strategy
> ---------------------------------------------------------------------------------------------
>
> Key: FLINK-40912
> URL: https://issues.apache.org/jira/browse/FLINK-40912
> Project: Flink
> Issue Type: Bug
> Components: Connectors / Kafka
> Reporter: Aleksandr Savonin
> Assignee: Aleksandr Savonin
> Priority: Major
> Labels: pull-request-available
>
> ExactlyOnceKafkaWriter lazily creates an AdminClient in getAdminClient() but
> never closes it. close() closes the current producer, the producer pool and
> the backchannel, and calls super.close(), but the adminClient field is not
> part of the closeAll(...) chain, and nothing else closes it.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
