You are viewing a plain text version of this content. The canonical link for it is here.
Posted to jira@arrow.apache.org by "Marton Elek (Jira)" <ji...@apache.org> on 2021/04/21 17:49:00 UTC

[jira] [Commented] (ARROW-12442) [CI] Set job timeouts on GitHub Actions

    [ https://issues.apache.org/jira/browse/ARROW-12442?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17326766#comment-17326766 ] 

Marton Elek commented on ARROW-12442:
-------------------------------------

Thanks [~apitrou] to open this issue, appreciate the effort. (FYI: We will do the same in Ozone https://issues.apache.org/jira/browse/HDDS-5131)

> [CI] Set job timeouts on GitHub Actions
> ---------------------------------------
>
>                 Key: ARROW-12442
>                 URL: https://issues.apache.org/jira/browse/ARROW-12442
>             Project: Apache Arrow
>          Issue Type: Task
>          Components: Continuous Integration
>            Reporter: Antoine Pitrou
>            Priority: Critical
>
> The default timeout for a single job in Github Actions is 6 hours:
> [https://docs.github.com/en/actions/reference/workflow-syntax-for-github-actions#jobsjob_idtimeout-minutes]
> While our jobs normally do not exceed 1 hour of runtime (and most of them are far quicker), sometimes some network issues may lead a job to take up the full 6 hours before timing out. Not only does this contribute to our own build queue growing unnecessarily, but it also impedes other Apache projects, since the number of jobs which can be run in parallel is capped at the organization level.
> We should therefore configure job timeouts which reflect our expectation of the overall runtime for each job. 1 hour should be a safe value for most of them, and would already dramatically reduce the impact of network issues.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)