Thanks Mathias and Gavin!<br><br>Just to add some additional insight into the Pika case, we're actually using heartbeat with the BlockingConnection.<br><br>There were a couple of minor bugs I had to fix in heartbeat. py, and yes, I hope to file bugs/pull requests.<br>
<br>With this added info, I can better explain the original issue. When we read a message, we may not ack it for hours. Since there's no other activity on the queue connection, the heartbeat.py logic never gets a chance to execute. Eventually the server times out our connection because it's sent frames but no Pika code gets run while we're working away.<br>
<br>Matt<br><br><div class="gmail_quote">On Tue, Jul 3, 2012 at 1:25 PM, Gavin M. Roy <span dir="ltr"><<a href="mailto:gmr@myyearbook.com" target="_blank">gmr@myyearbook.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div class="im"><br>
On Jul 3, 2012, at 4:18 PM, Matthias Radestock wrote:<br>
<br>
> On 03/07/12 21:14, Gavin M. Roy wrote:<br>
>> we're sending a Heartbeat frame and expecting the response.<br>
><br>
> Erhm. There is no response.<br>
<br>
</div>LOL well there you go, I'm going off memory (a bit rusty memory). In looking at the code, it's sending the frame on an IOLoop timer and looking for byte counter changes.<br>
<span class="HOEnZb"><font color="#888888"><br>
Gavin<br>
</font></span><div class="HOEnZb"><div class="h5">_______________________________________________<br>
rabbitmq-discuss mailing list<br>
<a href="mailto:rabbitmq-discuss@lists.rabbitmq.com">rabbitmq-discuss@lists.rabbitmq.com</a><br>
<a href="https://lists.rabbitmq.com/cgi-bin/mailman/listinfo/rabbitmq-discuss" target="_blank">https://lists.rabbitmq.com/cgi-bin/mailman/listinfo/rabbitmq-discuss</a><br>
</div></div></blockquote></div><br>