[
https://issues.apache.org/jira/browse/MAPREDUCE-7438?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Mudit Sharma updated MAPREDUCE-7438:
------------------------------------
Description:
Currently inactive nodes are removed from the Yarn local memory irrespective of
which state they are in. This makes the node removal process not too much
configurable
After this patch: https://issues.apache.org/jira/browse/YARN-10854
If autoscaling is enabled, lot many nodes go into DECOMMISSIONED state but
still other states like LOST, SHUTDOWN are very less and systems might want
them to be still visible on UI for better tracking
The proposal is to introduce a new config, which when set, will allow only
selective node states to be removed after going into untracked state.
Attaching PR for reference: https://github.com/apache/hadoop/pull/5680
Any thoughts/suggestions/feedbacks are welcome!
was:
Currently inactive nodes are removed from the Yarn local memory irrespective of
which state they are in. This makes the node removal process not too much
configurable
After this patch: https://issues.apache.org/jira/browse/YARN-10854
If autoscaling is enabled, lot many nodes go into DECOMMISSIONED state but
still other states like LOST, SHUTDOWN are very less and systems might want
them to be still visible on UI for better tracking
The proposal is to introduce a new config, which when set, will allow only
selective node states to be removed after going into untracked state.
Any thoughts/suggestions/feedbacks are welcome!
> Support removal of only selective node states in untracked removal flow
> -----------------------------------------------------------------------
>
> Key: MAPREDUCE-7438
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-7438
> Project: Hadoop Map/Reduce
> Issue Type: Improvement
> Reporter: Mudit Sharma
> Priority: Major
> Labels: pull-request-available
>
> Currently inactive nodes are removed from the Yarn local memory irrespective
> of which state they are in. This makes the node removal process not too much
> configurable
> After this patch: https://issues.apache.org/jira/browse/YARN-10854
> If autoscaling is enabled, lot many nodes go into DECOMMISSIONED state but
> still other states like LOST, SHUTDOWN are very less and systems might want
> them to be still visible on UI for better tracking
> The proposal is to introduce a new config, which when set, will allow only
> selective node states to be removed after going into untracked state.
>
> Attaching PR for reference: https://github.com/apache/hadoop/pull/5680
> Any thoughts/suggestions/feedbacks are welcome!
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: mapreduce-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: mapreduce-issues-h...@hadoop.apache.org
