[
https://issues.apache.org/jira/browse/YARN-3476?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14492593#comment-14492593
]
Rohith commented on YARN-3476:
------------------------------
bq. Not all of the application's logs were available in HDFS because it
encountered an error (token-related) trying to upload the logs.
Is this because of IllegalStateException caused failure?
There are 2 options
# do post-aggregation cleanup by handling IllegalStateException OR
# scheduler timer for those log directories which are not uploaded and which
are not deleted.
Thinking,does IllegalStateException is causing log not to be found in hdfs? If
it is not then I think simple way to handle is 1st approach.
> Nodemanager can fail to delete local logs if log aggregation fails
> ------------------------------------------------------------------
>
> Key: YARN-3476
> URL: https://issues.apache.org/jira/browse/YARN-3476
> Project: Hadoop YARN
> Issue Type: Bug
> Components: log-aggregation, nodemanager
> Affects Versions: 2.6.0
> Reporter: Jason Lowe
> Assignee: Rohith
>
> If log aggregation encounters an error trying to upload the file then the
> underlying TFile can throw an illegalstateexception which will bubble up
> through the top of the thread and prevent the application logs from being
> deleted.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
