[
https://issues.apache.org/jira/browse/MESOS-2790?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14569832#comment-14569832
]
Timothy Chen commented on MESOS-2790:
-------------------------------------
We actually recently put in a change, that capture the docker run error back to
the scheduler, so if you're using a custom scheduler it will be in your
TaskStatus update message.
> Capture and report docker issues as part of mesos-master logs.
> --------------------------------------------------------------
>
> Key: MESOS-2790
> URL: https://issues.apache.org/jira/browse/MESOS-2790
> Project: Mesos
> Issue Type: Improvement
> Components: docker
> Affects Versions: 0.22.1
> Environment: CentOS 7 x86-84
> Reporter: Casey Sybrandy
>
> I just ran into an issue today where docker ran into an error where it
> couldn't update it's iptables chain. (It was missing for some reason.) It
> took a while to debug the problem because as far as mesos was concerned,
> there were no errors. Executing the container simply failed.
> To try to make it easier to know what's going on, is it possible for the
> mesos slaves to capture any docker errors and report them as part of the
> mesos master log? Even better would be if it could report the error via the
> UI. At a bare minimum, and only if you can't nicely capture the docker
> error, report that a docker error has occurred and which machine it occurred
> on.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
