[
https://issues.apache.org/jira/browse/ARTEMIS-2537?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16974346#comment-16974346
]
Thomas Wood commented on ARTEMIS-2537:
--------------------------------------
Sorry, i deleted my previous reply(got my issues mixed up).
we have not seen this issue with NIO or MAPPED so far.
MAPPED seems to be the most stable for us.
will update if anything changes.
> AMQ222086: on replicated slave causes master to stop processing messages
> until slave is restarted.
> --------------------------------------------------------------------------------------------------
>
> Key: ARTEMIS-2537
> URL: https://issues.apache.org/jira/browse/ARTEMIS-2537
> Project: ActiveMQ Artemis
> Issue Type: Bug
> Affects Versions: 2.10.1
> Environment: * RHEL7
> * ASYNCIO on local storage
> Reporter: Thomas Wood
> Priority: Critical
>
> This seems to happen during journal compacting on slave but not 100% certain
> on that. This has happened several times during our testing and have only
> seen it when using ASYNCIO. We have switched to NIO to see if the problem
> happens again.
> {noformat}
> 2019-10-31 17:31:06,946 WARN [org.apache.activemq.artemis.core.server]
> AMQ222086: error handling packet PACKET(ReplicationAddMessage)[type=91,
> channelID=2, responseAsync=false, requiresResponse=false, correlationID=-1,
> packetObject=ReplicationAddMessage] for replication:
> java.lang.IllegalStateException: Cannot find add info 41041119 on compactor
> or current records
> at
> org.apache.activemq.artemis.core.journal.impl.JournalImpl.checkKnownRecordID(JournalImpl.java:1081)
> [artemis-journal-2.10.1.jar:2.10.1]
> {noformat}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
