[
https://issues.apache.org/jira/browse/YARN-7275?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
kartheek muthyala updated YARN-7275:
------------------------------------
Attachment: YARN-7275.005.patch
[~asuresh], Can you please review the following updated patch for YARN-7275.
It contains the following changes:
1. Introduced RECOVERY_COMPLETED, containerscheduler event for trying
queued recovered containers.
2. Introduced RECOVER_CONTAINER_PAUSED event in ContainerEvents for moving
the container that was recovered as paused from scheduled -> PAUSED after we
initialize the containerimpl object for it.
3. Reused the existing event –
ContainersLauncherEventType.RECOVER_CONTAINER_PAUSED for reacquiring paused
container, instead of failing.
+ earlier patch that you are aware of. The changes only take care of recovered
containers that were paused. For containers which has moved to SCHEDULED, but
didn’t get a chance to run to RUNNING, the state in which they will be
recovered is REQUESTED, hence ultimately they will be retried from the queue of
ContainerScheduler.
Let me know if the changes seems okay.
> NM Statestore cleanup for Container updates
> -------------------------------------------
>
> Key: YARN-7275
> URL: https://issues.apache.org/jira/browse/YARN-7275
> Project: Hadoop YARN
> Issue Type: Sub-task
> Reporter: Arun Suresh
> Assignee: kartheek muthyala
> Priority: Blocker
> Attachments: YARN-7275.001.patch, YARN-7275.002.patch,
> YARN-7275.003.patch, YARN-7275.004.patch, YARN-7275.005.patch
>
>
> Currently, only resource updates are recorded in the NM state store, we need
> to add ExecutionType updates as well.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
