FailoverTransport with MaxReconnectAttempts set does not recover quickly, when
used with ConnectionPooling....
--------------------------------------------------------------------------------------------------------------
Key: AMQ-1575
URL: https://issues.apache.org/activemq/browse/AMQ-1575
Project: ActiveMQ
Issue Type: Bug
Components: Transport
Affects Versions: 5.0.0
Environment: 5.1-SNAPSHOT
Reporter: Jason Rosenberg
This entry from the users forum describes the issue (and it includes a
reproducible test-case)...See the entry from me dated Dec 06, 2007.....
http://www.nabble.com/Timeout-and-Failover-on-a-queue-to13292887s2354.html#a14215834
I've confirmed that the issue happens with the latest 5.1-SNAPSHOT version, and
happens with either jencks AMQPool, or the built-in activeMQ PooledConnections.
The problem does not happen if no connection pooling is used.
Basically, it appears that the FailoverTransport is not properly communicating
with the TransportListener when brokers return to service.
Another user ('hbuch) has created a patch, implemented for 4.1.1, which seems
to solve the problem, which is also referred to in that same thread referred to
above....See entry from 'hbuch' dated Dec 05, 2007.....I've verified that that
fix seems to work (tested only under 4.1.1 so far)....
Anyway, it seems that this fix should be straightforward to fold back into the
source code for 5.1....
Please take a look.
Thanks,
Jason
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
