You are viewing a plain text version of this content. The canonical link for it is here.

Posted to notifications@accumulo.apache.org by "Josh Elser (JIRA)" <ji...@apache.org> on 2015/02/25 19:44:04 UTC

[jira] [Commented] (ACCUMULO-3625) IT's that restart processes should use a small zookeeper timeout

    [ https://issues.apache.org/jira/browse/ACCUMULO-3625?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14336947#comment-14336947 ] 

Josh Elser commented on ACCUMULO-3625:
--------------------------------------

Also need to balance this from being too small -- I've run into problems with tests that set 5s which isn't reliably enough to get a connection to ZooKeeper (on my testing platforms). Something like 15s would be my recommendation.

> IT's that restart processes should use a small zookeeper timeout
> ----------------------------------------------------------------
>
>                 Key: ACCUMULO-3625
>                 URL: https://issues.apache.org/jira/browse/ACCUMULO-3625
>             Project: Accumulo
>          Issue Type: Bug
>          Components: test
>            Reporter: Eric Newton
>            Priority: Trivial
>             Fix For: 1.7.0
>
>
> The default zookeeper timeout of 30s takes a long time to detect a process failure.  Make it shorter when we know we're going to be restarting services.



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