[
https://issues.apache.org/jira/browse/YARN-4696?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15154989#comment-15154989
]
Steve Loughran commented on YARN-4696:
--------------------------------------
This is another iteration with some other things needed to debug and harden the
code; probably best to let me iterate on this until I've got my spark-> ATS
work done, in case I find more problems. As addressed in patch 005, the scan
was having problems with 0 byte files. This may be because I'm using local fs,
not miniHDFS, this behaves differently.
I've also added more logging of what's going on, because there's enough async
operations going on in the client, it's a lot harder to debug why things aren't
working: is it there's been no flush client side, nothing picked up in the ATS,
etc, etc.
> EntityGroupFSTimelineStore to work in the absence of an RM
> ----------------------------------------------------------
>
> Key: YARN-4696
> URL: https://issues.apache.org/jira/browse/YARN-4696
> Project: Hadoop YARN
> Issue Type: Sub-task
> Components: timelineserver
> Affects Versions: 2.8.0
> Reporter: Steve Loughran
> Assignee: Steve Loughran
> Attachments: YARN-4696-001.patch, YARN-4696-002.patch,
> YARN-4696-003.patch, YARN-4696-005.patch
>
>
> {{EntityGroupFSTimelineStore}} now depends on an RM being up and running; the
> configuration pointing to it. This is a new change, and impacts testing where
> you have historically been able to test without an RM running.
> The sole purpose of the probe is to automatically determine if an app is
> running; it falls back to "unknown" if not. If the RM connection was
> optional, the "unknown" codepath could be called directly, relying on age of
> file as a metric of completion
> Options
> # add a flag to disable RM connect
> # skip automatically if RM not defined/set to 0.0.0.0
> # disable retries on yarn client IPC; if it fails, tag app as unknown.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
