Hello,
If you can see the admin console of all three brokers (I presume you mean the ActiveMQ web console), then you're not running in master slave. When using shared file system master/slave, only the master should start up fully, the other two slaves should not startup fully. One of the first things the broker tries to do at startup is gain a lock on the persistence adapter. If there is already one instance of ActiveMQ running, other instances configured to use the same ActiveMQ data folder, should not be able to obtain a lock and therefore don't start up. The output of the slaves should contain something like this: INFO | Database /opt/FUSE/AMQ/apache-activemq-5.5.0-fuse-00-53/data/kahadb/lock is locked... waiting 10 seconds for the database to be unlocked. Reason: java.io.IOException: File '/opt/FUSE/AMQ/apache-activemq-5.5.0-fuse-00-53/data/kahadb/lock' could not be locked. Slaves will not startup any transport connectors, network connectors, camel routes, jetty engine, whatsoever. If you don't see such behavior then your master slave isn't setup correctly. Does your Windows file system support file locking correctly? Regards, Torsten Mielke tors...@fusesource.com tmie...@blogspot.com
