I updated to RC3 and now get quite the opposite effect. if I send 100
messages and cause 20 to fail via a runtimeexception, messages (Im assuming
the ones that failed and shouldve been retried) get reprocessed over and
over, i.e. if I send 100 messages and the consumer writes them to a
database, I end up getting well over 100 entries in the database. This error
shows up many times in the logs - which definitely looks like a bug:
[ERROR] 2006-04-13 09:34:12,728 -
org.apache.activemq.ra.ServerSessionPoolImpl
(ServerSessionPoolImpl.java:178)- Error redispatching unconsumed messages
from stale session
java.lang.NullPointerException
at
org.apache.activemq.ra.ServerSessionPoolImpl.createServerSessionImpl(ServerSessionPoolImpl.java:60)
at
org.apache.activemq.ra.ServerSessionPoolImpl.getServerSession(ServerSessionPoolImpl.java:113)
at
org.apache.activemq.ra.ServerSessionPoolImpl.dispatchToSession(ServerSessionPoolImpl.java:135)
at
org.apache.activemq.ra.ServerSessionPoolImpl.removeFromPool(ServerSessionPoolImpl.java:175)
at
org.apache.activemq.ra.ServerSessionImpl.run(ServerSessionImpl.java:177)
at
org.apache.geronimo.connector.work.WorkerContext.run(WorkerContext.java:291)
at EDU.oswego.cs.dl.util.concurrent.PooledExecutor$Worker.run(Unknown
Source)
at java.lang.Thread.run(Thread.java:595)
--
View this message in context:
http://www.nabble.com/Transaction-Retry-Problems-%28AMQ4-RC3%29-t1406942.html#a3904140
Sent from the ActiveMQ - User forum at Nabble.com.
