<br><br><div><span class="gmail_quote">On 7/17/06, <b class="gmail_sendername">Vivek Khera</b> <<a href="mailto:vivek@khera.org">vivek@khera.org</a>> wrote:</span><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<br>On Jul 14, 2006, at 6:29 AM, Niels Huylebroeck wrote:<br><br>>> # in httpd.conf<br>>> KeepAlive On<br>> I've now also turned this on, thanks for the heads-up hadn't noticed<br>> this before (I run Centos
4.3 and it's disabled by default too)<br><br>You absolutely 100% don't want to do this on a busy public web site,<br>especially one behind a stateful firewall. Many web clients open up<br>multiple connections simultaneously and you then end up with a bunch
<br>of servers in keepalive state, where they could be helping others.</blockquote><div><br>I understand. In my case it isn't (and it's set to only 15 seconds), and the caching works well for the other 15 static files, so keepalive is not required in the general case with caching turned turned on.
<br></div><br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">for a small load internal application, you won't notice the difference.<br><br>
and if you have a stateful firewall, all those keepalive state<br>servers will suck up all the states on your firewall in no time flat.<br><br>the redhat folks are not totally wrong in disabling it by default.<br><br><br>
<br></blockquote></div><br><br clear="all"><br>-- <br>Mike Taht<br>PostCards From the Bleeding Edge<br><a href="http://the-edge.blogspot.com">http://the-edge.blogspot.com</a>