You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by GitBox <gi...@apache.org> on 2020/06/22 15:44:32 UTC

[GitHub] [flink] aljoscha commented on pull request #12157: [FLINK-17691][Connectors/Kafka] fix the bug FlinkKafkaProducer011 transactional.id too long when use Semantic.EXACTLY_ONCE mode

aljoscha commented on pull request #12157:
URL: https://github.com/apache/flink/pull/12157#issuecomment-647605986


   @freezhan Thanks a lot for analysing this bug! I think it would be better to limit the prefix instead of taking the md5 hash. The prefix potentially has some useful information that could help debugging problems with transactional IDs. I would suggest we instead cap the task name at some reasonable length before calling the `TransactionalIdsGenerator` constructor (otherwise we would also be cutting of the operator ID. And we should then also add a check inside the generator to verify that we really obey the length cap. What do you think?


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org