You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@mesos.apache.org by "Neil Conway (JIRA)" <ji...@apache.org> on 2016/10/06 21:42:21 UTC

[jira] [Commented] (MESOS-6228) Add timeout to /metrics/snapshot calls in tests

    [ https://issues.apache.org/jira/browse/MESOS-6228?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15553271#comment-15553271 ] 

Neil Conway commented on MESOS-6228:
------------------------------------

If we added a request timeout, the HTTP request would return successfully if fetching any metric times out. Not clear that this is actually better behavior. In this situation, we would {{VLOG(1)}} which metric has timed out; we could perhaps increase the verbosity of that error message and then enable the request timeout.

> Add timeout to /metrics/snapshot calls in tests
> -----------------------------------------------
>
>                 Key: MESOS-6228
>                 URL: https://issues.apache.org/jira/browse/MESOS-6228
>             Project: Mesos
>          Issue Type: Bug
>          Components: tests
>            Reporter: Neil Conway
>              Labels: mesosphere, newbie++
>
> In the unit tests, {{Metrics()}} does an {{http::get}} of the {{/metrics/snapshot}} endpoint. No {{timeout}} parameter is provided. That means if any metric cannot be fetched, the request hangs for 15 seconds and then dies with a mysterious / unclear error message. Digging into which metric has hung and for what reason requires a lot of time / debugging.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)