From: Nielsen, S. <SNi...@co...> - 2004-07-19 17:47:31
|
Cedric, There are no out of the ordinary messages in my log file. I downloaded = version 1.9.1. This version has proved to be much more stable thant the = latest (2.0.1).=20 Steve -----Original Message----- Le jeu 15/07/2004 =E0 16:51, Nielsen, Steve a =E9crit : > Hi, >=20 > I am testing middleman in a high proxy traffic environment (200 = req/sec and 90mbits/sec) and it is crashing every few minutes. I am looking for tip/pointers on = how to track down why. When it crashes the stack is being overwritten. =20 What are the last messages in your log ? You can have more log messages if you start middleman like this: % middleman -l /tmp/log -d 262144 =20 Regards, =20 -----Original Message----- From: Nielsen, Steve=20 Sent: Thursday, July 15, 2004 9:52 AM To: 'mid...@li...' Subject: middleman 2.0.1 crashing under high load Hi, I am testing middleman in a high proxy traffic environment (200 req/sec = and 90mbits/sec) and it is crashing every few minutes. I am looking for = tip/pointers on how to track down why. When it crashes the stack is = being overwritten. Here is what my setup is: Fedora Core 2 Latest middle-man from CVS I compiled with the following: --enable-debug --disable-ssl --disable-zlib --disable-ntlm I run it in gdb to try to get a backtrace. However the stack itself is = being corrupted. I was hoping there is a straightforward way for trapping the error to = find out where its occuring. Steve |