otterc edited a comment on pull request #31934: URL: https://github.com/apache/spark/pull/31934#issuecomment-805488580
> To confirm, for the example 2, I think the first 2 steps are not necessary, right? @Ngone51 I think the first 2 steps are necessary because in this edge case this can only happen when a stream is trying to merge a duplicate block, which was `stream1duplicate` in my example, and fails. The problem with that is that we were setting `isWriting=true` early in such cases. So when it fails then it can unset currentMapId. Let me know if I am missing some other cases. I can add a UTs for them as well. -- 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: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
