Use bonnie for test you storage space. And exucate "dmesg" (for Linux OS) and view any network problem.<br><br><br><div class="gmail_quote">On Sun, Feb 24, 2008 at 6:27 PM, Igor Sysoev <<a href="mailto:is@rambler-co.ru">is@rambler-co.ru</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;"><div><div></div><div class="Wj3C7c">On Sun, Feb 24, 2008 at 09:47:02AM +0000, nginx.mailinglist wrote:<br>
<br>
> Hi<br>
><br>
> thanks for all the replies! i should have provided more details but I was in<br>
> dispair and it was late at night, have clear head now after good sleep.<br>
><br>
> Ok heres info requested, the servers (i have another 6 pushing 120~mbit<br>
> average with lighttpd 1.4 compiled now, all servers same config)<br>
><br>
><br>
><br>
> OS: Suse 10.3 (minimal, i have barely anything installed and allmost<br>
> everythign compiled by me) <a href="http://pastebin.com/m41b5a046" target="_blank">http://pastebin.com/m41b5a046</a><br>
><br>
> Applications Running:<br>
> *nginx_0.5.35 (compiled <a href="http://pastebin.com/m6462475a" target="_blank">http://pastebin.com/m6462475a</a> ) used for file<br>
> downloads<br>
> *php_5.2.5 (compiled <a href="http://pastebin.com/m3c02ea6e" target="_blank">http://pastebin.com/m3c02ea6e</a> , ~20 threads started<br>
> with lighttps spawn fcgi), very short php scripts are run that use<br>
> X-Accell-Redirect to pass control to nginx of file serving<br>
> *lighttpd_1.5r2048 (compiled) used for file uploads, due to the progress<br>
> meter, fairly stable in production tho its from the svn<br>
> *custom php_cli socket deamons for inter server RPC, these are very light<br>
><br>
><br>
> *top <a href="http://pastebin.com/m2587b666*" target="_blank">http://pastebin.com/m2587b666*</a><br>
> *dstat <a href="http://pastebin.com/m2055efd4*" target="_blank">http://pastebin.com/m2055efd4*</a><br>
> *df <a href="http://pastebin.com/m724e52a7*" target="_blank">http://pastebin.com/m724e52a7*</a><br>
> *vmstat <a href="http://pastebin.com/m5cdc2f0b*" target="_blank">http://pastebin.com/m5cdc2f0b*</a><br>
> **<br>
> heres *lighttpd 1.4.18 config* from other servers on network (they are all<br>
> pretty much same just differing hosts) <a href="http://pastebin.com/m7b1af1e6" target="_blank">http://pastebin.com/m7b1af1e6</a><br>
><br>
><br>
> now the only thing that changed on this server is lighttpd 1.4.18 was<br>
> replaced by nginx, and now the request go thru nginx X-AcellRedirect not<br>
> lighttpd's mod_secdownload<br>
><br>
> the php5 fcgi scripts are very small it check file exists etc, then just<br>
> adds a downlaod id into headers for nginx to pick up and passes control onto<br>
> nginx the relevant bit is here > <a href="http://pastebin.com/m2bcbe7fd" target="_blank">http://pastebin.com/m2bcbe7fd</a><br>
><br>
> basically the server is capable of pushing above 100mbit easily before, now<br>
> the load is high, but this could be due to me setting high number of worker<br>
> threads like Igor said and more php-cgi processes floating around<br>
><br>
> so to summarise I need to figure out how to tweak this so i can move all the<br>
> servers from lighttpd to nginx for file downloads<br>
<br>
</div></div>You should try to set<br>
worker_processes 1;<br>
<br>
LA in Linux is sum of processes that run or ready to run and processes that<br>
wait for disk I/O. LA 6 and "70.5%wa" means that all 6 nginx workers wait<br>
for disk I/O. Here is your bottleneck.<br>
<div><div></div><div class="Wj3C7c"><br>
<br>
--<br>
Igor Sysoev<br>
<a href="http://sysoev.ru/en/" target="_blank">http://sysoev.ru/en/</a><br>
<br>
</div></div></blockquote></div><br><br clear="all"><br>-- <br>Please excuse my bad english...<br><br>С уважение,<br>Й. Георгиев.<br><br>WEB: <a href="http://gigavolt-bg.net/">http://gigavolt-bg.net/</a><br>Blog: <a href="http://live.gigavolt-bg.net/">http://live.gigavolt-bg.net/</a>