activemq-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From icfantv <adam.gor...@readytalk.com>
Subject JDBC master/slave runtime issue
Date Tue, 05 May 2009 19:11:15 GMT

We're running activeMQ in a master/slave configuration with a JDBC data store
and two brokers in a failover configuration.  According to the logs,
everything is happy.

The behavior we're seeing is that, in testing, when we kill the master the
slave takes over (as designed, and the log file shows this), but our java
code sporadically doesn't detect that the slave has taken over and the code
will just sit there blocking waiting (as if to be notified) that the broker
is back up.

What's really weird is that this is sporadic.  Sometimes it all works ok,
and sometimes it doesn't work.

Anyone have any ideas as to what might be going on?

We're running two brokers on the same server but on different ports (both
for the service and jetty) and things are essentially ok, save for the
apparent random failed reconnect.

Thanks,

--adam
-- 
View this message in context: http://www.nabble.com/JDBC-master-slave-runtime-issue-tp23394015p23394015.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.


Mime
View raw message