<html><head><meta http-equiv="Content-Type" content="text/html charset=iso-8859-1"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; "><div>Thanks Simon, Alvaro,</div><div><br></div><div>I don't think they are using big messages at least less that 2k.</div><div>I'll start looking at the logs, to see what I can find.</div><div><br></div><div>Thanks again</div><br><div apple-content-edited="true">
<div><div><b>Felipe Gutierrez<br></b></div><div><b>Sr. Consultant</b><br><i><span style="color: rgb(39, 78, 19); "><b>SpringSource/GoPivotal</b></span></i><br></div><div><br></div></div></div><div><div>On Oct 3, 2013, at 3:21 AM, Simon MacMullen <<a href="mailto:simon@rabbitmq.com">simon@rabbitmq.com</a>> wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite">On 03/10/2013 2:27AM, Alvaro Videla wrote:<br><blockquote type="cite">While I don't know about your exact problem keep in mind that a GC pause<br>could appear as a network partition.<br></blockquote><br>Err, what? Where did you get that idea?<br><br><blockquote type="cite">Also on some situation very big messages can cause network partitions<br>since Erlang nodes won't be able to ping each other.<br></blockquote><br>This is sadly still true. I had some idea that this was fixed in recent Erlang versions but I just tested with R16B01 and it's not.<br><br>It's pretty hard to run into though - you need to be sending huge enough messages that each message takes a multiple of net_ticktime (i.e. by default 60 seconds) to transfer.<br><br>I was only able to replicate it by building a cluster across a poor-quality wifi link. Which is obviously not recommended.<br><br>Still, even the theoretical possibility is disquieting, so I think a future RabbitMQ release will chunk large messages across clusters.<br><br>Cheers, Simon<br><br>-- <br>Simon MacMullen<br>RabbitMQ, Pivotal<br></blockquote></div><br></body></html>