[jboss-jira] [JBoss JIRA] Updated: (JGRP-1086) JChannel.close() should return only after channel is fully closed

[Sei Syvalta (JIRA)]

     [ https://jira.jboss.org/jira/browse/JGRP-1086?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Sei Syvalta updated JGRP-1086:
------------------------------

    Attachment: queue-close.patch


An experimental patch, which doesn't seem to make much difference...

Still after the patch a new node tries to connect to a node already disconnected in the same thread:

2009-11-03 13:51:45,332 [INFO ] [main      ] JgroupsTest    ] [5] channel-1 srv1.stop()
2009-11-03 13:51:50,848 [INFO ] [main      ] JgroupsTest    ] [6] srv1.start()
3.11.2009 13:51:50 INFO: JGroups version: 2.8.0.CR5

-------------------------------------------------------------------
GMS: address=channel-3, cluster=testClust, physical address=127.0.0.1:7800
-------------------------------------------------------------------
3.11.2009 13:51:53 WARNING: join(channel-3) sent to channel-1 timed out (after 3000 ms), retrying
3.11.2009 13:51:57 WARNING: join(channel-3) sent to channel-1 timed out (after 3000 ms), retrying
3.11.2009 13:51:58 WARNING: channel-1: GMS flush by coordinator failed
3.11.2009 13:52:00 WARNING: join(channel-3) sent to channel-1 timed out (after 3000 ms), retrying
3.11.2009 13:52:00 WARNING: channel-1: failed to collect all ACKs (expected=1) for view [channel-2|2] [channel-2] after 2000ms, missing ACKs from [channel-2]
3.11.2009 13:52:03 WARNING: join(channel-3) sent to channel-1 timed out (after 3000 ms), retrying


> JChannel.close() should return only after channel is fully closed
> -----------------------------------------------------------------
>
>                 Key: JGRP-1086
>                 URL: https://jira.jboss.org/jira/browse/JGRP-1086
>             Project: JGroups
>          Issue Type: Bug
>    Affects Versions: 2.8
>            Reporter: Sei Syvalta
>            Assignee: Bela Ban
>         Attachments: queue-close.patch
>
>
> JChannel.close() returns before actually waiting for all concurrent activity to be complete.

-- 
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