Varun Saxena commented on YARN-6736:

Final switch to v2 only would involve some kind of admin operation. We would 
have to switch off timeline server v1 for instance.
If AHS/ATSv1 server is running, for instance even though we have decided to 
switch to v2, client running the job may misconfigure v1 all across, get 
timeline delegation token from the server and the job would publish entities to 
timeline v1 while RM/NM would publish to v2. 
So, admin would need to shutdown ATSv1 server so that misconfiguration is 
caught at the time of job submission itself.

> Consider writing to both ats v1 & v2 from RM for smoother upgrades
> ------------------------------------------------------------------
>                 Key: YARN-6736
>                 URL: https://issues.apache.org/jira/browse/YARN-6736
>             Project: Hadoop YARN
>          Issue Type: Sub-task
>          Components: timelineserver
>            Reporter: Vrushali C
>            Assignee: Rohith Sharma K S
>         Attachments: YARN-6736-YARN-5355.001.patch
> When the cluster is being upgraded from atsv1 to v2, it may be good to have a 
> brief time period during which RM writes to both atsv1 and v2. This will help 
> frameworks like Tez migrate more smoothly. 

This message was sent by Atlassian JIRA

To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org

Reply via email to