[
https://issues.apache.org/jira/browse/AMQ-6797?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16138341#comment-16138341
]
Timothy Bish commented on AMQ-6797:
-----------------------------------
You'll need to provide a unit test that can reproduce the issue in order for
any investigation
> wss connections not being closed on ActiveMQ side
> -------------------------------------------------
>
> Key: AMQ-6797
> URL: https://issues.apache.org/jira/browse/AMQ-6797
> Project: ActiveMQ
> Issue Type: Bug
> Components: Transport
> Affects Versions: 5.14.0
> Environment: Debian (SMP Debian 3.16.7-ckt25-2+deb8u3 (2016-07-02)
> x86_64 GNU/Linux) on AWS, 4 cores, 7.5GB.
> Reporter: Marcos Moreno Martin
>
> We have clients that connect via wss to our ActiveMQ.
> Under some circumstances, we have socket connections that appear in the
> ActiveMQ web console, but when doing a netstat -an we can see that they are
> not open in the system.
> This means that ActiveMQ is enqueuing messages for that connections, but they
> are not there.
> This is the transport configuration:
> {code}
> <transportConnector name="wss"
> uri="wss://0.0.0.0:8082?maximumConnections=5000&wireFormat.maxFrameSize=104857600&needClientAuth=true&transport.enabledProtocols=TLSv1,TLSv1.1,TLSv1.2"
> allowLinkStealing="true" enableStatusMonitor="true" />
> {code}
> Is there a way of closing these connections? Is this an error solved in a
> newer release (could not find it in the release notes)? We have the
> enableStatusMonitor="true" option, but still the connection is not being
> closed on the ActiveMQ side. What we need to do right now is restarting
> ActiveMQ, but it is not desired.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)