yes it l b great if fair proxy balancer is added to nginx trunk<br><br><div class="gmail_quote">On Jan 31, 2008 2:31 PM, David Pratt <<a href="mailto:fairwinds@eastlink.ca">fairwinds@eastlink.ca</a>> 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 Grzegorz. This gives me a much better idea of what to expect. Thank<br>you for this. I am curious whether you have you done anything in the way<br>of comparing the effectiveness of the fair proxy balancer to other<br>balancing schemes like haproxy or lvm. Speed is a big factor for<br>
deployments so hoping speed will be good with the simplicity that this<br>option presents. Many thanks.<br><div class="Ih2E3d"><br>Regards,<br>David<br><br>Grzegorz Nosek wrote:<br></div><div><div></div><div class="Wj3C7c">
> On Thu, Jan 31, 2008 at 01:29:32PM -0400, David Pratt wrote:<br>>> Hi Grzegorz. I appreciate your explanation. It would be more convenient<br>>> to compile as an option since I am using an automated build process. If<br>
>> it is self contained, can you forsee any problems building with most<br>>> current 0.5.x branch or is this strictly 0.6.x? Also, what is the<br>>> request threshold that triggers the issue with round robin issue that I<br>
>> am aware. Many thanks.<br>><br>> The module works with 0.5.x as well as 0.6.x (if it doesn't work for<br>> you, please mail me with a bug report).<br>><br>> There's no threshold per se, it's just that the original load balancer<br>
> directs requests strictly round robin, i.e. 0-1-2-3-0-1-2-3 etc. This<br>> ensures that every backend gets the same number of requests.<br>><br>> upstream_fair always starts from backend 0 and works its way up until it<br>
> finds an idle peer (more or less). If your load effectively uses a<br>> single backend at one time, it'll always be backend 0. If it uses the<br>> power of two backends, they'll be 0 and 1 etc. Thus the first backend<br>
> will always have the most requests served, the second one will have more<br>> than the third etc.<br>><br>> Best regards,<br>> Grzegorz Nosek<br>><br>><br><br></div></div></blockquote></div><br>