AMQ 5.11.3 Old ACKed Messages Reappearing on Restart

[michaelcaplan]

Hello,

My AMQ server became unresponsive after throwing an OutOfMemoryError.  After
restarting, my queues (around a dozen of them) contained older messages (2 +
weeks old, thousands of messages) that I'm certain where ACKed.  The
earliest of these messages date back to the date I upgraded from AMQ 5.9.0
to AMQ 5.11.3 (two weeks ago).  

Putting aside the OutOfMemoryError exception, why would previously ACKed
messages (which where not showing up prior to the restart this morning) show
up after restart?

My install is configured with a journal persistence adaptor (journal log
files 5, datasource mysql 5.6).  Consumers are connecting over STOMP+NIO.

Some excerpts from my activemq.conf below.

Should I move to the non journaled persistence adaptor?


Thanks for your feedback,

Mike

        <persistenceFactory>
            <journalPersistenceAdapterFactory
dataDirectory="${activemq.base}/data" journalLogFiles="5"
dataSource="#mysql-ds"/>
        </persistenceFactory>


        <systemUsage>
            <systemUsage>
                <memoryUsage>
                    <memoryUsage percentOfJvmHeap="70" />
                </memoryUsage>
                <storeUsage>
                    <storeUsage limit="1 gb"/>
                </storeUsage>
                <tempUsage>
                    <tempUsage limit="100 mb"/>
                </tempUsage>
            </systemUsage>
        </systemUsage>

    <bean id="mysql-ds" class="org.apache.commons.dbcp.BasicDataSource"
destroy-method="close">
      <property name="driverClassName" value="com.mysql.jdbc.Driver"/>
      <property name="url"
value="jdbc:mysql://127.0.0.1/activemq?relaxAutoCommit=true"/>
      <property name="maxActive" value="200"/>
      <property name="poolPreparedStatements" value="true"/>
    </bean>




--
Sent from: http://activemq.2283324.n4.nabble.com/ActiveMQ-User-f2341805.html