mesos-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Chun-Hung Hsiao (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (MESOS-7801) Retry logic for unsuccessful `docker rm` during agent recovery
Date Mon, 28 Aug 2017 17:47:00 GMT

    [ https://issues.apache.org/jira/browse/MESOS-7801?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16144116#comment-16144116
] 

Chun-Hung Hsiao commented on MESOS-7801:
----------------------------------------

Hi [~xds2000]. This patch is an optimization for MESOS-7777, which has been landed a while
ago. I'd like to see this patch landed but since it is just an optimization, it might not
receive as high priority as other pending issues. I was wondering if you encounter any problem
that cannot be resolved by the patch for MESOS-7777. Could you provide more information to
help us understand the severity? Thanks!

> Retry logic for unsuccessful `docker rm` during agent recovery
> --------------------------------------------------------------
>
>                 Key: MESOS-7801
>                 URL: https://issues.apache.org/jira/browse/MESOS-7801
>             Project: Mesos
>          Issue Type: Improvement
>          Components: docker
>            Reporter: Chun-Hung Hsiao
>            Assignee: Chun-Hung Hsiao
>
> In MESOS-7777 we skip the failure when `docker rm` fails due to mount leakage during
agent recovery. In order not to leave residual docker containers in the docker daemon, we
could do a best-effort `docker rm` retry with an exponential backoff since we cannot control
when the leakage would be terminated.



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)

Mime
View raw message