Dave,<br>Each of those servers are 16G RAM, Quad Core, Dual Processor Machines by Dell. So - the memory shouldn't really be a problem.<br><br>Although someone did mention that <em>ip_conntrack</em> module might be a problem. <br>
<br>I am not sure what the issue is and not able to understand - why haproxy is unable to route requests.<br><br>Can you please tell me if there could be some other issue?<br><br>regards<br><br>Asif <br><br><br><div class="gmail_quote">
On Wed, Nov 19, 2008 at 6:43 PM, Dave Cheney <span dir="ltr"><<a href="mailto:dave@cheney.net">dave@cheney.net</a>></span> wrote:<br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
Hi Asif,<br>
<br>
What fails completely ? HaProxy, Mongrel or Nginx ?<br>
<br>
I noticed that <a href="http://192.168.1.16" target="_blank">192.168.1.16</a> for instance had over 50 mongrels assigned to it. At an average memory usage of 60-80 meg (just after startup) that is an enormous amount of memory in use (if I understand your config correctly), is the problem with the underlying server platform? Possibly swapping, or spending all its time scheduling processes?<br>
<br>
Cheers<br>
<br>
Dave<div><div></div><div class="Wj3C7c"><br>
<br>
On 19/11/2008, at 9:53 PM, Asif Ali wrote:<br>
<br>
</div></div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"><div><div></div><div class="Wj3C7c">
We have configured nginx + haproxy + mongrels [ more than 150 instances]. While it works perfectly fine for small configuration, a larger set of mongrels - fail completely.<br>
<br>
I have attached the configuration file for your reference. Apologies for the inconvenience if this is coming into the wrong list.<br>
<br>
Thanks in Advance<br>
<br>
Asif<br></div></div>
<haproxy.cfg><br>
</blockquote>
<br>
<br>
</blockquote></div><br>