docker-compose 搭建splash集群

目录结构

.
├── docker-compose.yaml
├── nginx
│   └── nginx.conf
└── script
    ├── run.sh
    └── splash_listener.py

docker-compose.yaml

version: "3"
services:
  nginx:
    image: nginx
    container_name: splash_nginx
    restart: always
    ports:
      - 8050:80
    volumes:
      - ./nginx/nginx.conf:/etc/nginx/nginx.conf
    depends_on:
      - splash1
      - splash2
  listener:
    image: python:3.8
    hostname: listener
    volumes:
      - ./script:/app
    command: bash /app/run.sh
    depends_on:
      - splash1
      - splash2
  splash1:
    hostname: splash1
    container_name: splash_1
    image: scrapinghub/splash
    restart: always
  splash2:
    hostname: splash2
    container_name: splash_2
    image: scrapinghub/splash
    restart: always

nginx.conf

user  nginx;
worker_processes  1;
error_log  /var/log/nginx/error.log warn;
pid        /var/run/nginx.pid;
events {
    worker_connections  1024;
}

http {
    include       /etc/nginx/mime.types;
    default_type  application/octet-stream;
    log_format  main  '$remote_addr - $remote_user [$time_local] "$request" '
                      '$status $body_bytes_sent "$http_referer" '
                      '"$http_user_agent" "$http_x_forwarded_for"';

    access_log  /var/log/nginx/access.log  main;    
    client_max_body_size 10m;
    sendfile        on;
    #tcp_nopush     on;
    keepalive_timeout  65;
    #gzip  on;

    upstream tomcat_client {
         server splash1:8050 weight=1;
         server splash2:8050 weight=1;
    }
    server {
        server_name "";
        listen 80 default_server;
        listen [::]:80 default_server ipv6only=on;
        location / {
            proxy_pass http://tomcat_client;
            proxy_redirect default;
            proxy_set_header Host $host;
            proxy_set_header X-Real-IP $remote_addr;
        }    
    }
}

splash_listener.py

# -*- coding:utf-8 -*-
# @Author: wmy
# @Time: 2020/7/3
# @Description:

import sys
import subprocess
import requests
import logging


class Listener(object):
    def __init__(self):
        self.splash_timeout = 10
        self.splash_ping_times = 2
        self.logger = self.get_logger()
        self.splash_servers = [
            {
                'host': 'splash1',
                'port': '8050',
                'name': 'splash_1',
            },
            {
                'host': 'splash2',
                'port': '8050',
                'name': 'splash_2',
            },
        ]

    def get_logger(self, name='splash_listener', level=logging.INFO):
        """
        获得一个logger
        :param name:
        :param level:
        :return:
        """
        logger = logging.getLogger(name)
        logger.setLevel(level)
        stream_handler = logging.StreamHandler()
        formatter = logging.Formatter('%(asctime)s - %(name)s - %(levelname)s: - %(message)s',
                                      datefmt='%Y-%m-%d %H:%M:%S')
        stream_handler.setFormatter(formatter)
        logger.addHandler(stream_handler)
        return logger

    def ping_splash(self, splash_host, splash_port):
        """
        尝试连接splash，测试splash服务是否正常
        :return: 正常，True,无法访问,False
        """
        splash_url = 'http://{}:{}'.format(splash_host, splash_port)
        try:
            resp = requests.get(splash_url, timeout=self.splash_timeout)
        except Exception as e:
            self.logger.error(u'请求出错.{}'.format(e))
            return False
        if resp.status_code != 200:
            self.logger.error(u'状态码异常.{}'.format(resp.status_code))
            return False
        return True

    def listen_splash(self):
        """
        监听splash，尝试连接splash，直到成功或者失败self.splash_ping_times次。
        :return: 成功，True，失败，False
        """
        for splash in self.splash_servers:
            mark = False
            # check splash
            for i in range(self.splash_ping_times):
                if self.ping_splash(splash['host'], splash['port']):
                    mark = False
                else:
                    mark = True
            if mark:
                # restart splash
                subprocess.Popen(args=['docker', 'restart', splash['name']])
                self.logger.error(u'{} splash服务异常，重启服务'.format(splash['name']))
            else:
                self.logger.info(u'{} splash服务正常'.format(splash['name']))


if __name__ == '__main__':
    import time

    while True:
        Listener().listen_splash()
        time.sleep(60*5)

run.sh

pip install -i https://pypi.tuna.tsinghua.edu.cn/simple requests
python /app/splash_listener.py