I'm using Apache & mod_wsgi to deploy a django application on several servers. I've read in multiple places (including this: http://blog.dscpl.com.au/2012/10/why-are-you-using-embedded-mode-of.html) that it's better to use the Daemon mode of wsgi. This would allow me to have control over process count and thread count per process, among other neat things :)
Now my is that I can have two or more instances of my django application on the same server (with each its own settings, databases, etc). For example:
Tough I think I understand how to use a different "process group" and "daemon process" configuration for multiple virtualhosts, i don't seem to wrap my mind around what i should do with multiple "sub-roots".
I run those under CentOS 6.2 distros. In the
/etc/httpd/conf.d/ directory I have one
.conf file for each instance which looks like this:
WSGIScriptAlias /prod-instance /opt/wsgi_applications/prod/app.wsgi
Should I use virtualhosts and have urls such as
http://prod-instance.team1-server/ instead? That would mean I should rely on the network managers to update the DNS tables, which is never fast enough for our clients. :)
I must admit I'm often lost when it comes to Apache configuration. Your help is welcome.