I'm using Django with mod_wsgi and need to tune the httpd.conf to handle 500 requests per second at peak. Right now I have the following configuration:
StartServers 3
ServerLimit 15
MinSpareThreads 75
MaxSpareThreads 250
MaxRequestWorkers 250
ThreadsPerChild 25
MaxMemFree 2048
MaxConnectionsPerChild 10000
WSGIApplicationGroup %{GLOBAL}
WSGIDaemonProcess django_process processes=23 threads=2
But I have no clue how it works and whether it is optimal configuration for my needs. Also, I don't know what each parameter means and how apache directives (StartServers etc.) work with mod_wsgi processes and threads.
Can anybody guide me how to tune this parameters or at least explain what do they mean, why do we need processes and threads and not just processes or threads?
Thanks is advance.
P.S. It is apache2/bin/httpd.worker Right now it handles around 200-300 requests per second, but the system is unstable. I got 502 errors. The server just stops.
