flink-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Till Rohrmann (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (FLINK-4319) Rework Cluster Management (FLIP-6)
Date Thu, 21 Sep 2017 23:36:01 GMT

    [ https://issues.apache.org/jira/browse/FLINK-4319?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16175658#comment-16175658

Till Rohrmann commented on FLINK-4319:

Hi [~suez1224],

thanks for testing out the current state of Flip-6. This is valuable feedback which we have
to address.

1) It's true that we don't wait on the completion of the {{JobMaster}} shutdown in the {{JobManagerRunner}}
yet. However, I thought that the {{JobClusterEntrypoint}} should call {{System.exit}} upon
termination of the job (at least eventually). This might point towards a bug.
2) The RM stops TMs after they were idle for a certain time (10s). However, the stopping logic
is not yet implemented for the {{YarnResourceManager}}. See YarnResourceManager.java:231.
This should be fairly straight forward. See FLINK-7076 for more information.
3) We are currently working on porting the web dashboard to the new architecture. See FLINK-7530
for the current state.

If you like, then you could take over FLINK-7076 since we didn't make much progress there.
Also FLINK-7530 could benefit from helping hands :-)

> Rework Cluster Management (FLIP-6)
> ----------------------------------
>                 Key: FLINK-4319
>                 URL: https://issues.apache.org/jira/browse/FLINK-4319
>             Project: Flink
>          Issue Type: Improvement
>          Components: Cluster Management
>    Affects Versions: 1.1.0
>            Reporter: Stephan Ewen
>            Assignee: Till Rohrmann
>              Labels: flip-6
> This is the root issue to track progress of the rework of cluster management (FLIP-6)

This message was sent by Atlassian JIRA

View raw message