Another bug related to this could be AMQ-1214. I see the same problem with a simple pair of brokers connected using a static NetworkConnector. When the receiving AQ server is down, the sender does the following every 5 seconds:
INFO [NetworkConnector] Establishing network connection between from vm://localhost?network=true to ssl://192.168.4.45:4081 WARN [NetworkConnector] Could not start network bridge between: vm://localhost?network=true and: ssl://192.168.4.45:4081 due to: java.net.ConnectException: Connection refused And it seems it leaks about 2 threads per minute in that state. Any idea if and where that is fixed in 5.0? I'd like to backport the fix if possible as I can't upgrade to 5.0 at this point. -- View this message in context: http://www.nabble.com/reconnection-to-unavailable-broker-causing-thread-memory-leak-tf4603206s2354.html#a13522716 Sent from the ActiveMQ - User mailing list archive at Nabble.com.
