Thanks for this FLIP, Yangze. Sorry for not involving in the previous discussion. In general I like the proposed direction to make related IDs have more rich information for debugging and correlation.
But I have a bit reminder for the changes. After failover restarting, every related IDs would be generated differently with before by random way, and I am not sure whether this was intentional design before, but it is the ground truth now. E.g. ExecutionAttemptID , and the ResultPartitionID is also derived from it to guarantee the uniqueness after failover. Based on current implementation, it seems that we will not store and rely on the previous ID states after failover, but I am not sure whether this assumption is valid for future features. Anyway, it is lucky that the proposed changes in this FLIP do not break the previous truth by introducing the `attemptNumber` in `ExecutionAttemptID`, so we do not need to further consider this issue now. +1 (binding). Best, Zhijiang ------------------------------------------------------------------ From:Till Rohrmann <trohrm...@apache.org> Send Time:2020 Apr. 16 (Thu.) 21:39 To:dev <dev@flink.apache.org> Subject:Re: [VOTE] FLIP-118: Improve Flink’s ID system Thanks for creating this FLIP, Yangze. +1 (binding). Cheers, Till On Thu, Apr 16, 2020 at 10:51 AM Yangze Guo <karma...@gmail.com> wrote: > Hi everyone, > > I'd like to start the vote of FLIP-118 [1], which improves the log > readability by adding more information to Flink’s IDs. This FLIP is > discussed in the thread[2]. > > The vote will be open for at least 72 hours. Unless there is an objection, > I will try to close it by April 21, 2020 10:00 UTC if we have received > sufficient votes. > > [1] > https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=148643521 > [2] > http://apache-flink-mailing-list-archive.1008284.n3.nabble.com/DISCUSS-FLIP-118-Improve-Flink-s-ID-system-td39321.html > > Best, > Yangze Guo >
