On Wed, 2009-06-10 at 01:43 -0700, dongabda wrote: > I'm testing the CMS SimpleAsyncConsumer example (last activemqcpp 3.0 RC > version). > I modified the brokerURI to test the failover transport, because it can > detect network fails. brokerURI="failover:(tcp://192.168.202.216:61616)"; > > When the network is down the CMS example application doesn't detect the > network failure. > This is the process: > 1. The producer (Other PC) sends a message when the networks fails. > 2. After some tieme (ex:1 min) the network is up again. > 3. When the network is up again, the CMS never receives any messages > (message of step1 and new messages) > > Any exception is captured, why? > > I did a JMS consumer to test the failover and it works fine. The brokerURI > is the same. brokerURI="failover:(tcp://192.168.202.216:61616)"; > > If the networks is down, after some time a message is showed describing the > problem. When the network is up again, it reconnect with the broker and > receives the messages. > > Why doesn't work on CMS? Will it work on CMS 3 final version?
Its most likely that there will be bugs in the Failover Transport in the 3.0 release, this is all new code and will take some time to mature. If you find cases where things aren't working for you the best thing to do is to create a new issue in Jira to document the problem providing as much information about you environment and if possible a complete sample application that demonstrates the problem. Regards Tim. -- Tim Bish http://fusesource.com http://timbish.blogspot.com/
