At Daylife, we're running a whole lotta Python code inside Apache using mod_python. This works great so long as each request can be processed quickly, and you don't need your Python processes to keep much of anything in RAM.
However, any real web app inevitably needs to break those rules, at which point you start to venture into a world of increasing pain. Thus arises a need for alternatives to our cozy, familiar, tried-and-true mod_python.
I've set up Lighttpd before, but this time I decided to try Nginx. (It's pronounced "engine x")
Here's a quick demo recipe for running two standalone Python FastCGI+WSGI servers that listen on Unix domain sockets, and configuring Nginx to load-balance between them. This config allows you to run more than one Python VM, but not need N VMs to handle N concurrent requests. It also allows you to restart VMs individually with no service downtime.
On CentOS x86_64, enable the EPEL repo and "sudo yum install python-flup nginx"
Download the two files in this directory:
Google Code – p16blog – python_nginx
"cp nginx.conf /etc/nginx".
Then: /etc/init.d/nginx restart
Start one Python FastCGI server (you downloaded this file above, right?)
python hello_flup.py -n 1 &
See that this creates a socket file in /tmp. Now hit any URL on your server, which is listening on port 88. Example: http://localhost:88/foo
Start a second server:
python hello_flup.py -n 2 &
Hit port 88 again. Hit refresh a few times. You should see the PID it reports alternating between two values. Woo, it's round-robining requests between those two servers! Now kill one server. You should see no interruption in service, but the PID value will no longer alternate between values.
Very nice. Now we can upgrade Python code on our server without interrupting service, or needing to restart Nginx itself. Possibly handy if you're running an API service. :)
Final note: you probably want as many VMs running as you have CPUs in your host. Eg. 4 VMs on a 4-way box. This allows you to avoid bottlenecking on the Python GIL.