[
https://issues.apache.org/jira/browse/FLINK-14363?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gary Yao closed FLINK-14363.
----------------------------
> Prevent vertex from being affected by outdated deployment (SchedulerNG)
> -----------------------------------------------------------------------
>
> Key: FLINK-14363
> URL: https://issues.apache.org/jira/browse/FLINK-14363
> Project: Flink
> Issue Type: Sub-task
> Components: Runtime / Coordination
> Affects Versions: 1.10.0
> Reporter: Zhu Zhu
> Assignee: Zhu Zhu
> Priority: Major
> Labels: pull-request-available
> Fix For: 1.10.0
>
> Time Spent: 20m
> Remaining Estimate: 0h
>
> DefaultScheduler currently will cancel latest execution of a vertex when the
> vertex version is outdated. This may lead to undesirable failover of a
> healthy vertex.
> I'd propose to remove the vertex cancel process in
> {{DefaultScheduler#stopDeployment}} because the cancellation is not not
> needed here, since the version of a scheduled vertex is only outdated after
> the vertex is canceled by others.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
