[
https://issues.apache.org/jira/browse/SPARK-30191?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Max Xie updated SPARK-30191:
-----------------------------
Description:
I run spark on yarn. I found that when driver lost its executors because of
machine hardware problem and all of service includes nodemanager, executor on
the node has been killed, it means that Resourcemanager can't update the
containers info on the node until Resourcemanager try to remove the node, but
it always takes 10 mins or longger, and in the meantime, AM don't add the new
resource request and driver missing the executors.
So maybe AM should add the factor `numExecutorsExiting` in YarnAllocator's
method `
updateResourceRequests` to optimize it.
was:
I run spark on yarn. I found that when driver lost its executors because of
machine hardware problem and all of service includes nodemanager, executor on
the node has killed, it means that Resourcemanager can't update the containers
info on the node until Resourcemanager try to remove the node, but it always
takes 10 mins or longger, and in the meantime, AM don't add the new resource
request and driver missing the executors.
So maybe AM should add the factor `numExecutorsExiting` in YarnAllocator's
method `
updateResourceRequests` to optimize it.
> AM should update pending resource request faster when driver lost executor
> --------------------------------------------------------------------------
>
> Key: SPARK-30191
> URL: https://issues.apache.org/jira/browse/SPARK-30191
> Project: Spark
> Issue Type: Improvement
> Components: YARN
> Affects Versions: 2.4.4
> Reporter: Max Xie
> Priority: Minor
>
> I run spark on yarn. I found that when driver lost its executors because of
> machine hardware problem and all of service includes nodemanager, executor on
> the node has been killed, it means that Resourcemanager can't update the
> containers info on the node until Resourcemanager try to remove the node,
> but it always takes 10 mins or longger, and in the meantime, AM don't add the
> new resource request and driver missing the executors.
> So maybe AM should add the factor `numExecutorsExiting` in YarnAllocator's
> method `
> updateResourceRequests` to optimize it.
>
>
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
