[jboss-jira] [JBoss JIRA] Commented: (JGRP-959) coord.stopFlush() and coord.close() called in sequence cause flush lockup

[Vladimir Blagojevic (JIRA)]

    [ https://jira.jboss.org/jira/browse/JGRP-959?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12464172#action_12464172 ] 

Vladimir Blagojevic commented on JGRP-959:
------------------------------------------

This issue is exhibited occasionally when channel.stopFlush() and channel#close() are called in sequence and channel is a group coordinator. Thread that invoked stopFlush() did not wait for a full completion of flush while it subsequently invoked close() which also triggered another flush (due to leave). However, this last flush immediately goes into wait-retry due to flush collision - previous flush did not finish yet. In meantime channel is torn down and a new flush triggered by close cannot finish properly - its channel does not exist any longer since it was already torn down by close(). 

> coord.stopFlush() and coord.close() called in sequence cause flush lockup
> -------------------------------------------------------------------------
>
>                 Key: JGRP-959
>                 URL: https://jira.jboss.org/jira/browse/JGRP-959
>             Project: JGroups
>          Issue Type: Bug
>    Affects Versions: 2.6.7, 2.6.8, 2.6.9
>            Reporter: Bela Ban
>            Assignee: Vladimir Blagojevic
>             Fix For: 2.6.10, 2.8
>
>         Attachments: JGroupsTest.java, JGroupsTest.java
>
>
> The attached program doesn't work

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira