You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@mesos.apache.org by "Benjamin Bannier (JIRA)" <ji...@apache.org> on 2019/05/01 19:46:00 UTC

[jira] [Commented] (MESOS-9738) Add per-framework metrics for offer round trip time.

    [ https://issues.apache.org/jira/browse/MESOS-9738?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16831215#comment-16831215 ] 

Benjamin Bannier commented on MESOS-9738:
-----------------------------------------

Thanks for filing this [~mzhu]. Like we discussed offline, we recently had a situation where we suspected problematic hoarding and offer handling, and exposing this directly would help to drill down into what could cause certain allocation issues. It is still not clear to me how an operator would make use of that information to deal with problematic, non-cooperative behavior though outside of changing framework weights to reduce the number of offers to such frameworks. I recently filed MESOS-9748 which proposes one automatic but only partial mitigation strategy, but I feel there might be more we could do here, even with existing APIs.

> Add per-framework metrics for offer round trip time.
> ----------------------------------------------------
>
>                 Key: MESOS-9738
>                 URL: https://issues.apache.org/jira/browse/MESOS-9738
>             Project: Mesos
>          Issue Type: Improvement
>          Components: allocation
>            Reporter: Meng Zhu
>            Priority: Major
>              Labels: mesosphere, resource-management
>
> This would provide more insights into framework responsiveness, help detect worrisome behaviors such as offer timeout, offer hoarding and etc.
> One tricky thing is that we need to take Mesos's own queuing delay into consideration.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)