Abimael-
A couple of things:
You don't use a networkConnector for master/slave. This is definitely
causing a problem. Network Connectors are to connect to *other*
master/slave pairs to provide scaling of traffic load. Remember:
Master/Slave *only* provides high availability for the messages in a
single data store.
Action 1: Remove <networkConnector /> config from both configs.
Your <destinationPolicy/> configuration is rather limiting. The
configuration of memoryLimit = 1mb means that each destination only gets
1 Mb of memory. This is a good default setting, but when you have a
busy destination, be sure to add an entry for the busy destination to
give it more than 1mb, or increase the default.
Action 2: After fixing and testing #1, look to tweak the default
memoryLimit, or add per-destination configs for busy queues/topics.
Hope this helps!
Matt Pavlovich
On 2/7/12 11:32 AM, Abimael wrote:
Any news ?
--
View this message in context:
http://activemq.2283324.n4.nabble.com/Activemq-Clustering-and-failover-tp4326849p4365514.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.
