[
https://issues.apache.org/jira/browse/YARN-4597?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Arun Suresh updated YARN-4597:
------------------------------
Attachment: YARN-4597.004.patch
Updated the patch based on [~jianhe]'s suggestions.
I've also updated the pull request. Attached the consolidated patch here to
kick of Jenkins.
bq. why do you need to add the additional check for SCHEDULED state ?
{code}
// Process running containers
if (remoteContainer.getState() == ContainerState.RUNNING ||
remoteContainer.getState() == ContainerState.SCHEDULED) {
{code}
This is actually needed. We need the RM to move the container out of the
ACQUIRED state. Currently, it happens only if container is in the RUNNING
state. We need to do so for SCHEDULED state as well.
> Add SCHEDULE to NM container lifecycle
> --------------------------------------
>
> Key: YARN-4597
> URL: https://issues.apache.org/jira/browse/YARN-4597
> Project: Hadoop YARN
> Issue Type: Bug
> Components: nodemanager
> Reporter: Chris Douglas
> Assignee: Arun Suresh
> Attachments: YARN-4597.001.patch, YARN-4597.002.patch,
> YARN-4597.003.patch, YARN-4597.004.patch
>
>
> Currently, the NM immediately launches containers after resource
> localization. Several features could be more cleanly implemented if the NM
> included a separate stage for reserving resources.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
