[
https://issues.apache.org/jira/browse/FLINK-17295?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17225972#comment-17225972
]
Stephan Ewen commented on FLINK-17295:
--------------------------------------
Let's keep also in mind that we might want to add reconciliation on master
failure at some point, meaning running execution attempts (tasks) scheduled by
previous leaders re-register at the current leader. That would speak against
making the fencing token part of this.
A random element of whichever kind makes sense to me. Either randomize the
job-id part, or add random entropy.
Adding random entropy and keeping the jobId makes things easier for sessions
with many jobs.
Can print/log the random part as well to help debugging.
> Refactor the ExecutionAttemptID to consist of ExecutionVertexID and
> attemptNumber
> ---------------------------------------------------------------------------------
>
> Key: FLINK-17295
> URL: https://issues.apache.org/jira/browse/FLINK-17295
> Project: Flink
> Issue Type: Sub-task
> Components: Runtime / Coordination
> Reporter: Yangze Guo
> Priority: Major
> Labels: pull-request-available
> Fix For: 1.12.0
>
>
> Make the ExecutionAttemptID being composed of (ExecutionVertexID,
> attemptNumber).
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
