I would recommend using haproxy if you have the budget for a separate box as a load balancer. As for doing it via the 2nd method... why would you want to do that? And the underlying assumption here - can nginx do that? How would it determine when the load has been reached for "this" box? (so that the rest become "additional" fastcgi requests)<br>
<br>-jf<br>
<br><br><div class="gmail_quote">On Thu, Sep 3, 2009 at 10:11 AM, Ilan Berkner <span dir="ltr"><<a href="mailto:iberkner@gmail.com">iberkner@gmail.com</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;">
So the time has come for us to add another web server (number 2) to our configuration to help with the amount of connections we're getting. I'm looking for some basic recommendations in terms of configuration of nginx.<div>
<br></div><div>That is:</div><div><br></div><div>1. Do I run exactly the same configuration on both boxes and load balance externally (i.e. nginx + php-fpm on each box + dedicated mysql server) or</div><div>2. Do I run nginx + php-fpm on box A and route additional fastCGI requests to box B?</div>
<div><br></div><div>What would the configuration look like? How do I preserver sessions? We are currently using memcached for session menagement and could place that on the dedicated data server. What's the "recommended" methodology?</div>
<div><br></div><div>Thanks!!!!!!!!</div>
</blockquote></div><br>