[
https://issues.apache.org/jira/browse/YARN-1776?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13941484#comment-13941484
]
Tsuyoshi OZAWA commented on YARN-1776:
--------------------------------------
[~zjshen], For FileSystemRMStateStore, I thought RM-HA is also supposed to work
with HDFS. Is this wrong? IMO, we can handle the case with temporary file as
follows:
1. Create UPDATING file, which includes both new delegation token information
and old delegation token information.
2. Create new delegation token file.
3. Delete obsolete delegation token file.
4. Delete UPDATING file.
At recovery time, if RM find UPDATING file, we can recover the state of
RMStateStore with UPDATING file.
> renewDelegationToken should survive RM failover
> -----------------------------------------------
>
> Key: YARN-1776
> URL: https://issues.apache.org/jira/browse/YARN-1776
> Project: Hadoop YARN
> Issue Type: Sub-task
> Reporter: Zhijie Shen
> Assignee: Zhijie Shen
> Attachments: YARN-1776.1.patch
>
>
> When a delegation token is renewed, two RMStateStore operations: 1) removing
> the old DT, and 2) storing the new DT will happen. If RM fails in between.
> There would be problem.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
