[
https://issues.apache.org/jira/browse/YARN-4606?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16453297#comment-16453297
]
Wangda Tan commented on YARN-4606:
----------------------------------
Thanks [~eepayne] / [~maniraj...@gmail.com],
Here's my understanding of the proposed approach:
1) When we compute {{max-am-resource-per-user}}, we uses #active-users +
#pending-users.
2) When we compute {{max-user-limit}}, we use #active-users only.
To me this is correct and (seems) same as what I proposed previously:
{code}
We should only consider a user is "active" if any of its application is active.
And CS will use the "#active-user-which-has-at-least-one-active-app" to compute
user-limit.
Computation of max-am-resource-per-user needs to be updated as well. We should
get a #users-which-has-pending-apps to compute max-am-resource-per-user.
{code}
I haven't checked very much details of the patch since [~maniraj...@gmail.com]
is working on update the tests, etc. Just one suggestion is: AppSchedulingInfo
is supports to cache status for pending resource, it might be better to avoid
invoking SchedulerAppAttempt's method from AppSchedulingInfo.
> CapacityScheduler: applications could get starved because computation of
> #activeUsers considers pending apps
> -------------------------------------------------------------------------------------------------------------
>
> Key: YARN-4606
> URL: https://issues.apache.org/jira/browse/YARN-4606
> Project: Hadoop YARN
> Issue Type: Bug
> Components: capacity scheduler, capacityscheduler
> Affects Versions: 2.8.0, 2.7.1
> Reporter: Karam Singh
> Assignee: Manikandan R
> Priority: Critical
> Attachments: YARN-4606.1.poc.patch, YARN-4606.POC.2.patch,
> YARN-4606.POC.patch
>
>
> Currently, if all applications belong to same user in LeafQueue are pending
> (caused by max-am-percent, etc.), ActiveUsersManager still considers the user
> is an active user. This could lead to starvation of active applications, for
> example:
> - App1(belongs to user1)/app2(belongs to user2) are active, app3(belongs to
> user3)/app4(belongs to user4) are pending
> - ActiveUsersManager returns #active-users=4
> - However, there're only two users (user1/user2) are able to allocate new
> resources. So computed user-limit-resource could be lower than expected.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
