[
https://issues.apache.org/jira/browse/YARN-6050?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15862132#comment-15862132
]
Robert Kanter commented on YARN-6050:
-------------------------------------
I was looking through the AM blacklisting code and thinking about this more,
and I think my previous solution isn't the right way to handle this. The AM
blacklist feature has a threshold where if you blacklist too many nodes, it
will unblacklist the nodes so you don't make the application unschedulable. So
it seems like the correct and consistent approach would be to change the
calculation here to take into account the number of nodes that are "eligible"
for scheduling the app, instead of just looking at all of the nodes. YARN-6148
was filed recently for a similar problem, but with node labels, and came to the
same conclusion.
I currently have a fix that does this, and am working on testing it. I should
hopefully have an updated patch early next week.
> AMs can't be scheduled on racks or nodes
> ----------------------------------------
>
> Key: YARN-6050
> URL: https://issues.apache.org/jira/browse/YARN-6050
> Project: Hadoop YARN
> Issue Type: Bug
> Affects Versions: 2.9.0, 3.0.0-alpha2
> Reporter: Robert Kanter
> Assignee: Robert Kanter
> Attachments: YARN-6050.001.patch, YARN-6050.002.patch,
> YARN-6050.003.patch, YARN-6050.004.patch, YARN-6050.005.patch,
> YARN-6050.006.patch
>
>
> Yarn itself supports rack/node aware scheduling for AMs; however, there
> currently are two problems:
> # To specify hard or soft rack/node requests, you have to specify more than
> one {{ResourceRequest}}. For example, if you want to schedule an AM only on
> "rackA", you have to create two {{ResourceRequest}}, like this:
> {code}
> ResourceRequest.newInstance(PRIORITY, ANY, CAPABILITY, NUM_CONTAINERS, false);
> ResourceRequest.newInstance(PRIORITY, "rackA", CAPABILITY, NUM_CONTAINERS,
> true);
> {code}
> The problem is that the Yarn API doesn't actually allow you to specify more
> than one {{ResourceRequest}} in the {{ApplicationSubmissionContext}}. The
> current behavior is to either build one from {{getResource}} or directly from
> {{getAMContainerResourceRequest}}, depending on if
> {{getAMContainerResourceRequest}} is null or not. We'll need to add a third
> method, say {{getAMContainerResourceRequests}}, which takes a list of
> {{ResourceRequest}} so that clients can specify the multiple resource
> requests.
> # There are some places where things are hardcoded to overwrite what the
> client specifies. These are pretty straightforward to fix.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
---------------------------------------------------------------------
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
