-
Bug
-
Resolution: Won't Do
-
Major
-
None
-
None
-
None
-
2
-
False
-
False
-
Undefined
-
If a broker is full (at globalMaxSize), management messages sent by the standard-controller component may fail in the following way:
2021-05-24T22:13:14.869Z WARN [server] AMQ222039: Messages sent to address 'ed889b45-6750-45d7-8c61-7aaf216c7977' are being dropped; size is currently: 0 bytes; max-size-bytes: -1; global-size-bytes: 67,110,168 2021-05-24T22:13:14.869Z WARN [AMQPSessionCallback] AMQ229102: Address "ed889b45-6750-45d7-8c61-7aaf216c7977" is full.: ActiveMQAddressFullException[errorType=ADDRESS_FULL message=AMQ229102: Address "ed889b45-6750-45d7-8c61-7aaf216c7977" is full.] at org.apache.activemq.artemis.core.paging.impl.PagingStoreImpl.page(PagingStoreImpl.java:812) at org.apache.activemq.artemis.core.persistence.impl.journal.AbstractJournalStorageManager.addToPage(AbstractJournalStorageManager.java:2235) at org.apache.activemq.artemis.core.postoffice.impl.PostOfficeImpl.processRoute(PostOfficeImpl.java:1509) at org.apache.activemq.artemis.core.postoffice.impl.PostOfficeImpl.route(PostOfficeImpl.java:1192) at org.apache.activemq.artemis.core.postoffice.impl.PostOfficeImpl.route(PostOfficeImpl.java:1064) at org.apache.activemq.artemis.core.server.impl.ServerSessionImpl.doSend(ServerSessionImpl.java:2163) at org.apache.activemq.artemis.core.server.impl.ServerSessionImpl.handleManagementMessage(ServerSessionImpl.java:2045) at org.apache.activemq.artemis.core.server.impl.ServerSessionImpl.send(ServerSessionImpl.java:1801) at org.apache.activemq.artemis.protocol.amqp.broker.AMQPSessionCallback.inSessionSend(AMQPSessionCallback.java:558) at org.apache.activemq.artemis.protocol.amqp.broker.AMQPSessionCallback.lambda$serverSend$2(AMQPSessionCallback.java:517) at org.apache.activemq.artemis.utils.actors.OrderedExecutor.doTask(OrderedExecutor.java:42) at org.apache.activemq.artemis.utils.actors.OrderedExecutor.doTask(OrderedExecutor.java:31) at org.apache.activemq.artemis.utils.actors.ProcessorBase.executePendingTasks(ProcessorBase.java:65) at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149) at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624) at org.apache.activemq.artemis.utils.ActiveMQThreadFactory$1.run(ActiveMQThreadFactory.java:118)
This causes error messages in the AMQ Online console reporting that the address is not configured on the broker. The message is in fact spurious, the address will be ready. If the message volumes on the broker is reduced so that the broker is no longer over max-address-size the condition will clear.
The root cause is https://issues.redhat.com/browse/ENTMQBR-2334 . The standard-controller could work around the issue in the same manner as the agent (in use a response queue prefixed by amq.management so that the broker never puts management responses to the store).
- relates to
-
ENTMQBR-2334 Management response messages are lost when Broker exceeds global-max-size limit and dynamic response queues are in use
- Closed