[
https://issues.apache.org/jira/browse/YARN-1372?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14095049#comment-14095049
]
Jian He commented on YARN-1372:
-------------------------------
bq. maybe we can just use NodeHeartbeatResponse#getContainersToCleanup to
notify NM to remove the containers from the context.
I was thinking to reuse getContainersToCleanup to notify NM to remove
containers. i.e. do not clean up the containers until gets pulled by AM. On a
second thought, this is not a good solution, please disregard this comment.
sorry for confusion.
Comments on the patch:
bq. Second patch uploaded that adds expiration to the entries in NM
How about tie the NM container lifecycle to the application itself ? i.e.
Clean all containers in context for each application in
response.getApplicationsToCleanup().
- DECOMMISSIONED/LOST state possible to receive the new event?
- In RMAppAttemptImpl, why add a new previousJustFinishedContainers?
- RMContainerImpl already has the nodeId, many changes related to adding the
NodeId are not needed. e.g. scheduler changes, RMContainerRecoverEvent changes.
- please add tests too.
> Ensure all completed containers are reported to the AMs across RM restart
> -------------------------------------------------------------------------
>
> Key: YARN-1372
> URL: https://issues.apache.org/jira/browse/YARN-1372
> Project: Hadoop YARN
> Issue Type: Sub-task
> Components: resourcemanager
> Reporter: Bikas Saha
> Assignee: Anubhav Dhoot
> Attachments: YARN-1372.prelim.patch, YARN-1372.prelim2.patch
>
>
> Currently the NM informs the RM about completed containers and then removes
> those containers from the RM notification list. The RM passes on that
> completed container information to the AM and the AM pulls this data. If the
> RM dies before the AM pulls this data then the AM may not be able to get this
> information again. To fix this, NM should maintain a separate list of such
> completed container notifications sent to the RM. After the AM has pulled the
> containers from the RM then the RM will inform the NM about it and the NM can
> remove the completed container from the new list. Upon re-register with the
> RM (after RM restart) the NM should send the entire list of completed
> containers to the RM along with any other containers that completed while the
> RM was dead. This ensures that the RM can inform the AM's about all completed
> containers. Some container completions may be reported more than once since
> the AM may have pulled the container but the RM may die before notifying the
> NM about the pull.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
