I'm getting a strange error w/ ActiveMQ 5.10.0 as well as 5.11.1 (upgrading
didn't help). The logs state that the temp store is full but at 0%:
javax.jms.ResourceAllocationException: Temp Store is Full (0% of 2147483648)
Restarting ActiveMQ always clears it up but eventually we get this error
again. The admin console shows that the queue is well over 100% full though
- usually reaching 200%-600%. However, on the physical machines our disk
usage is normal w/ plenty of space and our queues and topics show all items
dequeued. Our usage of ActiveMQ is frequent small messages and the behavior
that we're seeing is the temp store will jump from 0% to >200% usage w/ one
message of ~300 bytes and stay at that level until we restart that one
instance.
Our setup is 3 AMQ instances clustered using ZooKeeper and the Replicated
LevelDB. Here's the config for our usage limits:
<systemUsage>
<systemUsage sendFailIfNoSpaceAfterTimeout="10000">
<memoryUsage>
<memoryUsage limit="4 gb"/>
</memoryUsage>
<storeUsage>
<storeUsage limit="20 gb"/>
</storeUsage>
<tempUsage>
<tempUsage limit="2 gb"/>
</tempUsage>
</systemUsage>
</systemUsage>
And we have turned producer flow control off for all queues and topics.
We've been dealing with this for a few months now and am a bit stuck as to
what to try next. Any help would be greatly appreciated!
--
View this message in context:
http://activemq.2283324.n4.nabble.com/Temp-Store-is-Full-False-Positive-tp4695836.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.
