I recently tried upgrading from Hazelcast 2.0.3 to Hazelcast 2.1. Using my unmodified HZ-2.0.3 configuration, my 3-node system can no longer establish a connection with each other to form a cluster.
The following message repeats indefinitely:
2012-05-02 21:58:52,869 T=[main] WARN ~ [fe80:0:0:0:1031:3bff:fe12:cd71]:10000 [prod] Failed to connect, node joined= false, allConnected= false to all other members after 0 seconds. L=[TcpIpJoinerOverAWS]
2012-05-02 21:58:52,870 T=[main] WARN ~ [fe80:0:0:0:1031:3bff:fe12:cd71]:10000 [prod] Rebooting after 10 seconds. L=[TcpIpJoinerOverAWS]
Are there new configuration options with Hazelcast 2.1 that I should be aware of? I have since reverted to 2.0.3 to get my cluster up and running again.