hmm. I see we already have an issue for the RemoteJMXBrokerFacade at https://issues.apache.org/jira/browse/AMQ-2785 vote it up I guess or if you can, submit a patch. All contributions most welcome.
On 5 September 2012 12:00, Gary Tully <gary.tu...@gmail.com> wrote: > There is no easy solution for this at the moment. > > Since https://issues.apache.org/jira/browse/AMQ-2306 both the master > and slave expose them selves over JMX so both urls will be valid. > The difference is that only one will report isSlave==false > > There is a jmx facade used by the web console that could be useful if > your client has java code, but it needs some additional smarts to be > aware of the isSlave attribute on a broker. > > Currently it is happy if it finds a broker MBean. > > Have a peek at: > org.apache.activemq.web.RemoteJMXBrokerFacade#createConnection in [1] > > Seems like that facade needs to be able to iterate through the list of > urls or restrict access to just the master. I think it would make a > sensible enhancement. > > [1] > http://svn.apache.org/viewvc/activemq/trunk/activemq-web/src/main/java/org/apache/activemq/web/RemoteJMXBrokerFacade.java?view=markup > > On 5 September 2012 06:18, Gaurav Sharma <gaurav.cs.sha...@gmail.com> wrote: >> For the JDBC Master Slave config, just like there exists the Failover >> Transport for the broker url, what is the best way to specify the JMX >> hosts in a properties file: >> jmxUrl=service:jmx:rmi:///jndi/rmi://<host1>:1616/jmxrmi, \ >> service:jmx:rmi:///jndi/rmi://<host2>:1616/jmxrmi, \ >> service:jmx:rmi:///jndi/rmi://<host3>:1616/jmxrmi >> >> so that I don't have to worry about managing 3 MBeanServerConnection's >> in my code and then trying to figure which one is the master or which >> one is alive, re-election, etc? The alternative of managing the three >> references and error-handling, is just messier. >> >> Thanks in advance for the help. >> >> -gaurav > > > > -- > http://fusesource.com > http://blog.garytully.com -- http://fusesource.com http://blog.garytully.com
