[jboss-jira] [JBoss JIRA] Created: (JGRP-679) Member propagates non-existing coordinator
Member propagates non-existing coordinator
------------------------------------------
Key: JGRP-679
URL: http://jira.jboss.com/jira/browse/JGRP-679
Project: JGroups
Issue Type: Bug
Affects Versions: 2.2.8
Environment: Oracle Application Server Cluster running application using JGroups
Reporter: Peter Fuchs
Assigned To: Bela Ban
Priority: Critical
Our application is running on a clustered application server environment (Oracle AS 10gR2)
with 9 application instances in total. We are using JGroups for keeping the object caches
of all application instances in sync.
From time to time we encounter the problem that our application cannot
be started and seems to "hang". After setting the JGroups logger to DEBUG we
could see that it was JGroups that was trying to connect to a coordinator that did not
exist anymore. After some investigation we found out that there was one running instance
propagating a coordinator that did not exist any longer. The only solution in that case
was to restart the instance that was propagating the wrong coordinator address.
As the version we currently use (2.2.8) is quite old I checked Jira for fixes regarding
this issue but could not really identify this specific behaviour (maybe due to my poor
knowledge about the internals of JGroups). We are planning to use JGroups V2.4.1(SP4) in
our future software releases (as we are bound to JDK 1.4.2) but would like to know if this
issue has been fixed in that version.
Please find attached an example log file excerpt.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
http://jira.jboss.com/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira