[jira] [Commented] (ARTEMIS-4797) Failover connection references are not always cleaned up in NettyAcceptor, leaking memory

["Erwin Dondorp (Jira)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/ARTEMIS-4797?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17852677#comment-17852677 ] 

Erwin Dondorp commented on ARTEMIS-4797:
----------------------------------------

[~Josh B]
I'm trying to explain the 'ghost' connections between cluster nodes that are created in my cluster(s) during startup. as reported in ARTEMIS-3157. This is somewhat similar to this report, though it is not a growing problem in my case.
 * Are the old connections in your case still visible in the "Browse Connections" screen?
 * Do they look different than 'good' connections?

> Failover connection references are not always cleaned up in NettyAcceptor, leaking memory
> -----------------------------------------------------------------------------------------
>
>                 Key: ARTEMIS-4797
>                 URL: https://issues.apache.org/jira/browse/ARTEMIS-4797
>             Project: ActiveMQ Artemis
>          Issue Type: Bug
>          Components: OpenWire
>            Reporter: Josh Byster
>            Priority: Major
>          Time Spent: 20m
>  Remaining Estimate: 0h
>
> I'm still trying to parse through exactly what conditions this occurs in, since I'm able to reproduce it in a very specific production setup but not in an isolated environment locally.
> For context, we have custom slow consumer detection that closes connection IDs with slow consumers. These connections are connected via failover transport using client ActiveMQ Classic 5.16.4 (OpenWire). This seems to be specific to Netty.
> It appears this specific order of events causes the connection to not get cleaned up and retained indefinitely on the broker. With frequent kicking of connections, this ends up causing the broker to eventually OOM.
> 1. Connection is created, {{ActiveMQServerChannelHandler}} is created as well
> 2. {{ActiveMQServerChannelHandler#createConnection}} is called, {{active}} flag is set {{true}}.
> 3. A few minutes go by, then we call {{ActiveMQServerControl#closeConnectionWithID}} with the connection ID.
> 4. {{ActiveMQChannelHandler#exceptionCaught}} gets called—*this is the key point that causes issues*. The connection is cleaned up if and only if this is *not* called. The root cause of the exception is {{AbstractChannel.close(ChannelPromise)}}, however the comment above it says this is normal for failover.
> 5. The {{active}} flag is set to {{false}}.
> 6. {{ActiveMQChannelHandler#channelInactive}} gets called, but does *not* call {{listener.connectionDestroyed}} since the {{active}} flag is false.
> 7. The connection is never removed from the {{connections}} map in {{NettyAcceptor}}, causing a leak and eventual OOM of the broker if it happens frequently enough.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@activemq.apache.org
For additional commands, e-mail: issues-help@activemq.apache.org
For further information, visit: https://activemq.apache.org/contact