You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@qpid.apache.org by "Martin Ritchie (JIRA)" <qp...@incubator.apache.org> on 2009/09/28 15:58:16 UTC
[jira] Commented: (QPID-2033) QpidTestCase does not ensure failed
brokers are fully stopped.
[ https://issues.apache.org/jira/browse/QPID-2033?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12760250#action_12760250 ]
Martin Ritchie commented on QPID-2033:
--------------------------------------
QTC does perform a Process.destroy() which should be equivellent to a Kill -9.
That said we really should have a graceful way of stopping the external brokers.
> QpidTestCase does not ensure failed brokers are fully stopped.
> --------------------------------------------------------------
>
> Key: QPID-2033
> URL: https://issues.apache.org/jira/browse/QPID-2033
> Project: Qpid
> Issue Type: Bug
> Components: Java Tests
> Reporter: Martin Ritchie
> Assignee: Martin Ritchie
>
> When an external broker fails to become ready within the alloted 30 seconds QpidTestCase simply runs the clean script and logs the error.
> In the case of the Java broker it is possible that the process has started up and bound to the Management port then failed in a hung state. As a result QTC does not see the READY statement and assumes the broker has died.
> Yet it does not kill the process to ensure it is really gone.
> This results in all subsequent test runs failing with 'broker failed to be come ready' errors and in a CI situation all test runs will stop as the rogue process will not be terminated.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
---------------------------------------------------------------------
Apache Qpid - AMQP Messaging Implementation
Project: http://qpid.apache.org
Use/Interact: mailto:dev-subscribe@qpid.apache.org