Hi, I'm writing some tests to replicate activemq running out of persistence storage. To do this, I am simply testing some sending and receiving messages, then filling up a test queue with persistent messages, purge the queue via JMX, then trying to send and receive messages again to see if all recovers.
The problem I am having is that purging the queue does not seem to clear the persistent store. Even after purging the filled up queue of all messages, AMQ is still in a state where it won't process further messages, with it just printing the following message in the logs: INFO | Usage(activeMQMessageStore:queue://test.out:store) percentUsage=64%, usage=76438840, limit=67108864, percentUsageMinDelta=1%;Parent:Usage(activeMQMessageStore) percentUsage=113%, usage=76438840, limit=67108864, percentUsageMinDelta=1%: Persistent store is Full, 100% of 67108864. Stopping producer (ID:gl02720m.gamesys.corp-49306-1382002892643-1:43:1:1) to prevent flooding queue://test.out. See http://activemq.apache.org/producer-flow-control.html for more info (blocking for: 667s) I am using the kahadb persistence adapter and I've read in a few posts that this stores various metadata in the store rather than just the messages. My question is, how can I recover from this situation? Or at least test the recover? At the minute, the only way I seem to be able to clear the space is to delete the data directory for kaha. For my testing my settings are as follows: <systemUsage> <systemUsage> <memoryUsage> <memoryUsage limit="10 mb"/> </memoryUsage> <storeUsage> <storeUsage limit="64mb" name="activeMQMessageStore"/> </storeUsage> <tempUsage> <tempUsage limit="1 mb"/> </tempUsage> </systemUsage> </systemUsage> Thanks! -- View this message in context: http://activemq.2283324.n4.nabble.com/Recovering-from-running-out-of-persistence-storage-testing-tp4672927.html Sent from the ActiveMQ - User mailing list archive at Nabble.com.
