[
https://issues.apache.org/jira/browse/YARN-5018?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Li Lu updated YARN-5018:
------------------------
Attachment: YARN-5018-YARN-2928.001.patch
OK let me do a first-aid style fix to this problem. The problem occurs because
we tried to launch the first aggregation call immediately after the collector
got started. However, some context information is not set until postPut. This
would cause the aggregation method sees null values within the context.
This fix is a "first-aid" style fix because I simply introduced a wait (15
secs) between the start of a collector and the first aggregation call. A more
comprehensive fix would introduce a callback function from the collector
manager to collectors so that we can tell the collectors all context data
should be ready. Let me know if you have a strong preference on that one.
Thanks!
> Online aggregation logic should not run immediately after collectors got
> started
> --------------------------------------------------------------------------------
>
> Key: YARN-5018
> URL: https://issues.apache.org/jira/browse/YARN-5018
> Project: Hadoop YARN
> Issue Type: Sub-task
> Components: timelineserver
> Reporter: Li Lu
> Assignee: Li Lu
> Labels: yarn-2928-1st-milestone
> Attachments: YARN-5018-YARN-2928.001.patch
>
>
> In app level collector, we launch the aggregation logic immediately after the
> collector got started. However, at this time, important context data has yet
> to be published to the container. Also, if the aggregation result is empty,
> we do not need to publish them.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
