<div>Just thinking about it quickly, it's always possible it has nothing to do with Asterisk. There are many instances where I run into issues with a poorly configured servers when they have even a little bump in HTTP traffic. This was years ago though, and it was an issue to do with a web server and not Asterisk, but look into your kernel's configuration. Sometimes the kernel's settings are setup for a normal USER and not designed to handle the memory allocation a server demands. The fix for me back then was something to do with the MAXIMUM PAGE REQUESTS or SIZE maybe. Basicly the kernel couldn't keep track of all the HTTP processes.
</div>
<div> </div>
<div>Now that I'm reading this over I doubt it's your problem because Asterisk doesn't fork. But while we're at it, tell me a bit more about your system. What operating system (and version)? The problem could also be with your method of load generation, but I wouldn't know that since I've never tried load testing a system.
</div>
<div> </div>
<div>Lastly, I know FreeBSD started incorporating a basic DDoS protection a few years back and maybe that's also in some of these newer Linux distros. They would detect a flood and start to limit the bandwidth. These are just ideas, I don't really like any of them.
</div>
<div> </div>
<div>Sometimes the kernel will report issues to SYSLOGD. Might want to check your error and message logs.</div>
<div> </div>
<div>cat /proc/meminfo</div>
<div> </div>
<div>On a Linux box will give you memory limits and how close you are to them. They're not exactly what I was looking for, but maybe that will help. All TCP connections require the Kernel to page the information but I can't seem to find out how to access that limit if any.
<br><br> </div>
<div><span class="gmail_quote">On 9/20/07, <b class="gmail_sendername">Wai Wu</b> <<a href="mailto:wkwu@calltrol.com">wkwu@calltrol.com</a>> wrote:</span>
<blockquote class="gmail_quote" style="PADDING-LEFT: 1ex; MARGIN: 0px 0px 0px 0.8ex; BORDER-LEFT: #ccc 1px solid"><br>Hi everyone,<br><br>I am running into wall today with simultaneous call limits. I have two<br>Asterisk machines (fast 3GHz C2D with 2GB of ram). I tried to create a
<br>lot of sip calls from one machine to the other by issuing AMI Originate<br>commands to one machine. The machine that makes calls plays a message<br>(demo-intruct) upon the other machine answer. The machine receives the
<br>calls just waits for 40 seconds then hangs up. Throught the manager<br>connection, I was creating 10 calls per-second. I also have sip phone<br>registered with the calling machine. At around 150 to 200 calls. When I<br>
call the machine that's making all the calls, most of the calls couldn't<br>go through. For the ones that went through, most of them will drop off<br>within seconds of the call. But here is catch. When I run 'top', the cpu
<br>is idling 97%. My question is. Is there a limit on the number of<br>simultaneous calls Asterisk can handle? I know I have very fast systems.<br>Shouldn't they be able to handle that many calls? What is your take?<br>
<br>Thnx<br><br>_______________________________________________<br><br>Sign up now for AstriCon 2007! September 25-28th. <a href="http://www.astricon.net/">http://www.astricon.net/</a><br><br>--Bandwidth and Colocation Provided by
<a href="http://www.api-digital.com--">http://www.api-digital.com--</a><br><br>asterisk-users mailing list<br>To UNSUBSCRIBE or update options visit:<br> <a href="http://lists.digium.com/mailman/listinfo/asterisk-users">
http://lists.digium.com/mailman/listinfo/asterisk-users</a><br></blockquote></div><br><br clear="all"><br>-- <br>/Nick