[
https://issues.apache.org/jira/browse/YARN-8018?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16401130#comment-16401130
]
Eric Yang commented on YARN-8018:
---------------------------------
[~csingh] RM rest API is multi-threaded. There is no guarantee that request
will be handled via ServiceClient in ordered fashion. It would be best to have
a memory queue of size of n (default to 1) in AM for queuing the requests, and
commit service definition after flex is completed. If AM goes down, then AM
recovery will use the last copy on HDFS. User will need to resubmit the
upgrade request again if there was something in the memory queue prior to AM
crash. This will also prevent getting stuck in queued request. This also
reduce complexity of logic that we run in RM.
> Yarn service: Add support for initiating service upgrade
> --------------------------------------------------------
>
> Key: YARN-8018
> URL: https://issues.apache.org/jira/browse/YARN-8018
> Project: Hadoop YARN
> Issue Type: Sub-task
> Reporter: Chandni Singh
> Assignee: Chandni Singh
> Priority: Major
> Attachments: YARN-8018.wip.patch
>
>
> Add support for initiating service upgrade which includes the following main
> changes:
> # Service API to initiate upgrade
> # Persist service version on hdfs
> # Start the upgraded version of service
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
