[
https://issues.apache.org/jira/browse/YARN-5774?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15688458#comment-15688458
]
Daniel Templeton commented on YARN-5774:
----------------------------------------
Thanks for making those changes. A few more comments:
* The patch needs a rebase.
* In {{DominantResourceCalculator.normalize()}}, would it make more sense to
only replace the zero part of the step size with that part of the minimum, e.g.
only the CPU if the memory is fine?
* I kinda want to make {{NormalizableRequest}} an abstract base class instead
of an interface. The fact that the requests are asking for resources is core
to what they do. I would also like to find a better name, because the fact
that the things can be normalized is incidental. The main point is that they
request resources. Unfortunately {{ResourceRequest}} is already taken. Maybe
{{ContainerRequest}}? Maybe rename {{ResourceRequest}} to {{ContainerRequest}}
and reuse {{ResourceRequest}}? [~kasha], any opinion here?
* {{NormalizableRequest}} could use a little more explanation in the class
javadoc and some method javadocs.
* {{ResourceRequest}} and {{UpdateContainerRequest}} should have
{{getCapacity()}} and {{setCapacity()}} set as {{Override}}
* In {{TestResourceCalculator.testNormalize()}}, it would be nice to test
normalization to a value that is more than the minimum, e.g. min=2, max=8,
ask=3.
> MR Job stuck in ACCEPTED status without any progress in Fair Scheduler if set
> yarn.scheduler.minimum-allocation-mb to 0.
> ------------------------------------------------------------------------------------------------------------------------
>
> Key: YARN-5774
> URL: https://issues.apache.org/jira/browse/YARN-5774
> Project: Hadoop YARN
> Issue Type: Bug
> Components: resourcemanager
> Affects Versions: 3.0.0-alpha1
> Reporter: Yufei Gu
> Assignee: Yufei Gu
> Labels: oct16-easy
> Attachments: YARN-5774.001.patch, YARN-5774.002.patch,
> YARN-5774.003.patch, YARN-5774.004.patch, YARN-5774.005.patch,
> YARN-5774.006.patch
>
>
> MR Job stuck in ACCEPTED status without any progress in Fair Scheduler
> because there is no resource request for the AM. This happened when you
> configure {{yarn.scheduler.minimum-allocation-mb}} to zero.
> The problem is in the code used by both Capacity Scheduler and Fair
> Scheduler. {{scheduler.increment-allocation-mb}} is a concept in FS, but not
> CS. So the common code in class RMAppManager passes the
> {{yarn.scheduler.minimum-allocation-mb}} as incremental one because there is
> no incremental one for CS when it tried to normalize the resource requests.
> {code}
> SchedulerUtils.normalizeRequest(amReq, scheduler.getResourceCalculator(),
> scheduler.getClusterResource(),
> scheduler.getMinimumResourceCapability(),
> scheduler.getMaximumResourceCapability(),
> scheduler.getMinimumResourceCapability()); --> incrementResource
> should be passed here.
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
