You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by "Zhu Zhu (JIRA)" <ji...@apache.org> on 2019/01/31 02:22:00 UTC
[jira] [Commented] (FLINK-11471) Job is reported RUNNING
prematurely for queued scheduling
[ https://issues.apache.org/jira/browse/FLINK-11471?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16756792#comment-16756792 ]
Zhu Zhu commented on FLINK-11471:
---------------------------------
Hi [~yanghua] and [~NicoK], do you have the solution for this issue already?
For batch job, there can always be some tasks running, some tasks in scheduling and some tasks pending for scheduling checking. What would the job status be in this case?
> Job is reported RUNNING prematurely for queued scheduling
> ---------------------------------------------------------
>
> Key: FLINK-11471
> URL: https://issues.apache.org/jira/browse/FLINK-11471
> Project: Flink
> Issue Type: Bug
> Components: Distributed Coordination, Scheduler
> Affects Versions: 1.7.1
> Reporter: Nico Kruber
> Assignee: vinoyang
> Priority: Major
>
> With queued scheduling enabled (seems to be the default now), the job's status is changed to RUNNING before all tasks are actually running.
> Although {{JobStatus#RUNNING}} states
> {quote}Some tasks are scheduled or running, some may be pending, some may be finished.{quote}
> you may argue whether this is the right thing to report, e.g. in the REST interface, when a user wants to react on the actual state change from SCHEDULED to RUNNING. It seems, some intermediate state is missing here which would clarify things.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)