Exact command to use gunicorn with pypy

user_1411410 0
Viewed 1174

I've searched Google but can't seem to find an example of the command to use pypy with gunicorn. My current command to start gunicorn looks like this:

gunicorn = subprocess.Popen(['gunicorn', 
                             '-k', 'gevent', 
                             '-b', 'unix:/dev/shm/gunicorn.sock', 
                             '-w', '4', 
                             'wsgi:app'])

I have tried the following:

gunicorn = subprocess.Popen(['pypy', 'gunicorn', 
                             '-k', 'gevent', 
                             '-b', 'unix:/dev/shm/gunicorn.sock', 
                             '-w', '4', 
                             'wsgi:app'])

and

gunicorn = subprocess.Popen(['gunicorn', 
                             '-k', 'gevent', 
                             '-b', 'unix:/dev/shm/gunicorn.sock', 
                             '-w', '4', 
                             'pypy', 'wsgi:app'])

but both did not work.

So my question is, how exactly can I start my app (wsgi:app) using pypy within the gunicorn command?

pypy gunicorn python
2 Answers

In Mint, this appears to work:

mkdir api_test && cd $_
apt install pypy3 siege
pypy3 -m venv .venv_pypy3
source .venv_pypy3/bin/activate
pypy3 -m pip install gunicorn
nano gunicorn_test.py

Paste:

import multiprocessing

import gunicorn.app.base


def number_of_workers():
    return (multiprocessing.cpu_count() * 2) + 1


def handler_app(environ, start_response):
    response_body = b'Works fine'
    status = '200 OK'

    response_headers = [
        ('Content-Type', 'text/plain'),
    ]

    start_response(status, response_headers)

    return [response_body]


class StandaloneApplication(gunicorn.app.base.BaseApplication):

    def __init__(self, app, options=None):
        self.options = options or {}
        self.application = app
        super().__init__()

    def load_config(self):
        config = {key: value for key, value in self.options.items()
                  if key in self.cfg.settings and value is not None}
        for key, value in config.items():
            self.cfg.set(key.lower(), value)

    def load(self):
        return self.application


if __name__ == '__main__':
    options = {
        'bind': '%s:%s' % ('127.0.0.1', '8080'),
        'workers': number_of_workers(),
    }
    StandaloneApplication(handler_app, options).run()

Run:

pypy3 gunicorn_test.py

Benchmark in new terminal for client:

$ siege --benchmark --time=20s --concurrent=20 http://127.0.0.1:8080
** SIEGE 4.0.4
** Preparing 20 concurrent users for battle.
The server is now under siege...
Lifting the server siege...
Transactions:              85638 hits
Availability:             100.00 %
Elapsed time:              19.30 secs
Data transferred:           0.82 MB
Response time:              0.00 secs
Transaction rate:        4437.20 trans/sec
Throughput:             0.04 MB/sec
Concurrency:               19.66
Successful transactions:       85638
Failed transactions:               0
Longest transaction:            0.14
Shortest transaction:           0.00

Note that siege will hang here if using too high a concurrency! 200 was no problem in uvicorn on pypy3. I'm testing best of 3 in a VirtualBox on Windows 10 using 75% memory according to Process Hacker. CPython appears to be less demanding:

Python3 setup in server terminal:

deactivate
apt install python3-venv
python3 -m venv .venv_python3
source .venv_python3/bin/activate
python3 -m pip install gunicorn
python3 gunicorn_test.py

Benchmark in (new) client terminal:

$ siege --benchmark --time=20s --concurrent=20 http://127.0.0.1:8080
** SIEGE 4.0.4
** Preparing 20 concurrent users for battle.
The server is now under siege...
Lifting the server siege...
Transactions:              65340 hits
Availability:             100.00 %
Elapsed time:              19.02 secs
Data transferred:           0.62 MB
Response time:              0.01 secs
Transaction rate:        3435.33 trans/sec
Throughput:             0.03 MB/sec
Concurrency:               19.71
Successful transactions:       65340
Failed transactions:               0
Longest transaction:            0.16
Shortest transaction:           0.00
user_8194170

Gunicorn stays in the folder Scripts, so it depends. Make sure when you make gunicorn available global by using os environment, it stays in Pypy's subfolder.

user_137821510
Related