geode-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "ASF subversion and git services (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (GEODE-857) SystemFailure.stopWatchdog will always wait 1 second
Date Thu, 17 Mar 2016 21:45:33 GMT

    [ https://issues.apache.org/jira/browse/GEODE-857?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15200449#comment-15200449
] 

ASF subversion and git services commented on GEODE-857:
-------------------------------------------------------

Commit ac3d3b4c5294bbb64a54fef191584c6d4306de04 in incubator-geode's branch refs/heads/develop
from [~upthewaterspout]
[ https://git-wip-us.apache.org/repos/asf?p=incubator-geode.git;h=ac3d3b4 ]

GEODE-857: Wait for watchdog threads to start in SystemFailureJUnitTest

I can't reproduce this failure, but maybe it's possible the isAlive flag
is not set immediately on all platforms. So we'll have the test wait for
the threads to be alive, rather than assert.


> SystemFailure.stopWatchdog will always wait 1 second
> ----------------------------------------------------
>
>                 Key: GEODE-857
>                 URL: https://issues.apache.org/jira/browse/GEODE-857
>             Project: Geode
>          Issue Type: Bug
>            Reporter: Dan Smith
>            Assignee: Dan Smith
>             Fix For: 1.0.0-incubating.M2
>
>
> This method is called during distributed system shutdown. It synchronizes on failureSync,
and then calls Thread.join on the watchDog thread and waits one second.
> The watchDog thread can't wake up until it can get the failureSync, so the Thread.join
call will always end up waiting.
> This adds a lot of time to our tests, because we often disconnect the distributed system.



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

Mime
View raw message