Tell us more about "they're no longer
communicating to each other."  In a JMX viewer, for a particular queue
(pick one, doesn't matter which) on each broker, which consumers are
present on that queue before and after the restart?

Also, what's in the logs of the brokers immediately before, during, and
after the firewall shutdown?

Tim
On Feb 29, 2016 2:56 PM, "yang.yang.zz" <yang.yang...@outlook.com> wrote:

> Hi:
>
> I'm running into an massive message blocking with ActiveMQ 5.13.1. This
> issue is found during our reliance testing. This testing is set due to a
> similar issue found in the field, where our customer is using 5.11.1 which
> have encountered similar disconnection issues.
>
> Here's our topology:
>
>                   Consumer0
>                         |
>                     broker0
>
>           /            |              \
>          /      (network)          \
>         /              |                 \
>   broker1        broker2      ...   broker17
>
>   producer1     producer2   ...  producer17
>
>
> The issue we found here is, if, at broker0 side, if we set the firewall to
> drop all the incoming messages from broker1 to broker17 for longer than 5
> mins. Then remove the firewall setting. Then even we observed the
> connection
> between broker0 and broker1-17 gets re-established. But they're no longer
> communicating to each other. We have to restart the broker0 to get the
> communication back.
>
> Here're the abstracted configurations:
>
> *broker0:*
>
>         <transportConnectors>
>             <transportConnector name="openwire" uri="tcp://0.0.0.0:61616
> "/>
>         </transportConnectors>
>
>
> *broker1-17:*
>
>        <networkConnectors>
>                   <networkConnector name="broker0"
> uri="static:(tcp://broker0:61616)" duplex="true" />
>         </networkConnectors>
>       <transportConnectors>
>             <transportConnector name="openwire"
> uri="tcp://0.0.0.0:61616
> maximumConnections=100&amp;wireFormat.maxFrameSize=104857600"/>
>         </transportConnectors>
>
>
> Btw, this issue is not seen with fewer producers. say less than 5
> producers.
> But this issue is 100% reproducible with 17 producers.
>
> I spent hours and tried different transportConnector and networkConnector
> configuration options but it didn't help. Is this a configuration mistake
> or
> a bug with 5.13.1?
>
> Many thanks,
> -Yang
>
>
>
>
>
>
>
> --
> View this message in context:
> http://activemq.2283324.n4.nabble.com/5-13-1-message-blocked-tp4708592.html
> Sent from the ActiveMQ - User mailing list archive at Nabble.com.
>

Reply via email to