Gunicorn 部署 #

┌─────────────────────────────────────────────────────────────┐
│                    Gunicorn 优势                             │
├─────────────────────────────────────────────────────────────┤
│                                                             │
│   Uvicorn:                                                  │
│   - 单进程                                                  │
│   - 无法利用多核 CPU                                        │
│   - 进程崩溃需要手动重启                                     │
│                                                             │
│   Gunicorn + Uvicorn:                                       │
│   - 多进程管理                                              │
│   - 充分利用多核 CPU                                        │
│   - 自动重启崩溃进程                                        │
│   - 优雅重启                                                │
│   - 进程监控                                                │
│                                                             │
└─────────────────────────────────────────────────────────────┘

pip install gunicorn uvicorn[standard]

gunicorn main:app -w 4 -k uvicorn.workers.UvicornWorker -b 0.0.0.0:8000

import multiprocessing

bind = '0.0.0.0:8000'
workers = multiprocessing.cpu_count() * 2 + 1
worker_class = 'uvicorn.workers.UvicornWorker'
keepalive = 120
timeout = 120

gunicorn main:app -c gunicorn.conf.py

import multiprocessing
import os

bind = '0.0.0.0:8000'

workers = int(os.getenv('WORKERS', multiprocessing.cpu_count() * 2 + 1))

worker_class = 'uvicorn.workers.UvicornWorker'

worker_connections = 1000

timeout = 120

keepalive = 5

threads = 1

max_requests = 1000
max_requests_jitter = 50

preload_app = True

accesslog = '-'
errorlog = '-'
loglevel = 'info'

def on_starting(server):
    print('Starting Gunicorn server')

def on_exit(server):
    print('Shutting down Gunicorn server')

def worker_exit(server, worker):
    print(f'Worker {worker.pid} exited')

def when_ready(server):
    print('Server is ready')

import multiprocessing

workers = multiprocessing.cpu_count() * 2 + 1

preload_app = True

max_requests = 1000
max_requests_jitter = 50

accesslog = '-'
errorlog = '-'
loglevel = 'info'

access_log_format = '%(h)s %(l)s %(u)s %(t)s "%(r)s" %(s)s %(b)s "%(f)s" "%(a)s" %(D)s'

accesslog = '/var/log/gunicorn/access.log'
errorlog = '/var/log/gunicorn/error.log'

[Unit]
Description=FastAPI Application
After=network.target

[Service]
User=www-data
Group=www-data
WorkingDirectory=/var/www/my-api
Environment="PATH=/var/www/my-api/venv/bin"
ExecStart=/var/www/my-api/venv/bin/gunicorn main:app -c gunicorn.conf.py
ExecReload=/bin/kill -s HUP $MAINPID
ExecStop=/bin/kill -s TERM $MAINPID
Restart=always
RestartSec=5

[Install]
WantedBy=multi-user.target

sudo systemctl start my-api
sudo systemctl stop my-api
sudo systemctl restart my-api
sudo systemctl status my-api
sudo systemctl enable my-api

upstream fastapi {
    server 127.0.0.1:8000;
    keepalive 32;
}

server {
    listen 80;
    server_name example.com;

    location / {
        proxy_pass http://fastapi;
        proxy_http_version 1.1;
        proxy_set_header Host $host;
        proxy_set_header X-Real-IP $remote_addr;
        proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
        proxy_set_header X-Forwarded-Proto $scheme;
        proxy_set_header Connection "";
        proxy_connect_timeout 60s;
        proxy_send_timeout 60s;
        proxy_read_timeout 60s;
    }
}

upstream fastapi {
    least_conn;
    server 127.0.0.1:8000;
    server 127.0.0.1:8001;
    server 127.0.0.1:8002;
    keepalive 32;
}

workers = (2 * CPU_CORES) + 1

worker_connections = 1000

timeout = 120
graceful_timeout = 30
keepalive = 5

threads = 2
workers = multiprocessing.cpu_count()

import multiprocessing
import os

bind = os.getenv('BIND', '0.0.0.0:8000')

workers = int(os.getenv('WORKERS', multiprocessing.cpu_count() * 2 + 1))

worker_class = 'uvicorn.workers.UvicornWorker'

worker_connections = 1000

timeout = int(os.getenv('TIMEOUT', 120))

graceful_timeout = 30

keepalive = 5

threads = int(os.getenv('THREADS', 1))

max_requests = 1000
max_requests_jitter = 50

preload_app = True

accesslog = os.getenv('ACCESS_LOG', '-')
errorlog = os.getenv('ERROR_LOG', '-')
loglevel = os.getenv('LOG_LEVEL', 'info')

access_log_format = '%(h)s %(l)s %(u)s %(t)s "%(r)s" %(s)s %(b)s "%(f)s" "%(a)s" %(D)s'

def on_starting(server):
    server.log.info('Starting Gunicorn server')

def when_ready(server):
    server.log.info('Server is ready. Spawning workers')

def worker_int(worker):
    worker.log.info(f'Worker {worker.pid} received INT or QUIT signal')

def worker_abort(worker):
    worker.log.info(f'Worker {worker.pid} received SIGABRT signal')

#!/bin/bash

export WORKERS=4
export TIMEOUT=120
export LOG_LEVEL=info

gunicorn app.main:app -c gunicorn.conf.py