[
https://issues.apache.org/jira/browse/YARN-10229?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17144552#comment-17144552
]
Young Chen commented on YARN-10229:
-----------------------------------
Thanks for the patch [~BilwaST]!
What do you think about moving the federation check & checkIfAppExists calls to
inside AMRMProxyService as well? Then ContainerManagerImpl won't need to care
whether Federation is enabled, etc.
I'm trying to understand the use case for AMRMProxy being enabled for. Here's
the patch that added that: https://issues.apache.org/jira/browse/YARN-5646 It
may be worth thinking about the ways distr. scheduling and federation can be
combined here - I suspect if distr. scheduling is enabled we may not want to
entirely skip processing for non-federated applications..
Let me know how the proposed changes sound - in the meantime I'll try to
understand the interaction AMRMProxy has with distr. scheduling.
> [Federation] Client should be able to submit application to RM directly using
> normal client conf
> ------------------------------------------------------------------------------------------------
>
> Key: YARN-10229
> URL: https://issues.apache.org/jira/browse/YARN-10229
> Project: Hadoop YARN
> Issue Type: Bug
> Components: amrmproxy, federation
> Affects Versions: 3.1.1
> Reporter: JohnsonGuo
> Assignee: Bilwa S T
> Priority: Major
> Attachments: YARN-10229.001.patch, YARN-10229.002.patch,
> YARN-10229.003.patch, YARN-10229.004.patch, YARN-10229.005.patch
>
>
> Scenario： When enable the yarn federation feature with multi yarn clusters,
> one can submit their job to yarn-router by *modified* their client
> configuration with yarn router address.
> But if one still wants to submit their jobs via the original client (before
> enable federation) to RM directly, it will encounter the AMRMToken exception.
> That means once enable federation ,if some one want to submit job, they have
> to modify the client conf.
>
> one possible solution for this Scenario is:
> In NodeManger, when the client ApplicationMaster request comes:
> * get the client job.xml from HDFS "".
> * parse the "yarn.resourcemanager.scheduler.address" parameter in job.xml
> * if the value of the parameter is "localhost:8049"(AMRM address),then do
> the AMRMToken valid process
> * if the value of the parameter is "rm:port"(rm address),then skip the
> AMRMToken valid process
>
>
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
