redhat 上 supervisor 的安装和使用

1. 安装

可以通过pip 安装, 这种方式安装对python的版本要求比较严格, 我在写文档的时候, supervisor==3.3.5, Supervisor is intended to work on Python 3 version 3.4 or later and on Python 2 version 2.7.

pip install supervisor

如果是内网机没有网的话可以去Github仓库, clone 一份源码来安装.

python setup.py install 

安装过程中可能会报缺少meld3 python 安装包, 安装之后就可以正常supervisor了

2. 使用

如果系统上有多个版本的python, 直接运行 supervisord 会显示 command not found, 这是因为supervisor 安装在某个版本的python 目录下了, 需要手动找到, 创建软链接.

find / -name supervisord
//2.7 显示是
/usr/local/python2.7/bin/supervisord
//3.7 显示是
/root/.pyenv/versions/venv370/bin/supervisord

//创建链接
ln -s /usr/local/python2.7/bin/supervisord  /usr/local/bin/supervisord
//或者:
ln -s /root/.pyenv/versions/venv370/bin/supervisord  /usr/bin/supervisord

//同理:
ln -s /usr/local/python2.7/bin/supervisorctl  /usr/local/bin/supervisorctl
//或者:
ln -s /root/.pyenv/versions/venv370/bin/supervisorctl  /usr/bin/supervisorctl

ln -s /usr/local/python2.7/bin/echo_supervisord_conf  /usr/local/bin/echo_supervisord_conf
//或者:
ln -s /root/.pyenv/versions/venv370/bin/echo_supervisord_conf  /usr/bin/echo_supervisord_conf

现在可以使用了.

• 配置

  创建配置文件, 根据官网, 我将目录创建在/etc目录下

  echo_supervisord_conf > /etc/supervisord.conf
  

  打开文件, 在文件末尾追加:

  vi /etc/supervisord.conf
  
  //我这里监控一个我的python3.7 的uwsgi进程, 这个进程用来启动我的django 项目
  
  [program:dbyw-python3]
  directory=/dbtool/python_software/dbyw-python3
  command=uwsgi --ini uwsgi.ini
  autostart=true
  stdout_logfile=/etc/supervisor.d/log/dbyw3_stdout.log
  stderr_logfile=/etc/supervisor.d/log/dbyw3_stderr.log
      
  //监控python celery
  [program:dbyw-python3-celery]
  directory=/dbtool/python_software/dbyw-python3
  command=celery -A celery_tasks worker --loglevel=info --logfile celery_beat.log
  autostart=true
  stdout_logfile=/etc/supervisor.d/log/dbyw3_celery_stdout.log
  stderr_logfile=/etc/supervisor.d/log/dbyw3_celery_stderr.log
  
  
  //监控zookeeper
  [program:zookeeper]
  command=/dbtool/python_software/kafka/bin/zookeeper-server-start.sh config/zookeeper.properties
  autostart=true
  stdout_logfile=/etc/supervisor.d/log/zookeeper_stdout.log
  stderr_logfile=/etc/supervisor.d/log/zookeeper_stderr.log
  
  
  //监控kafka
  [program:kafka]
  command=/dbtool/python_software/kafka/bin/kafka-server-start.sh config/server.properties
  autostart=true
  stdout_logfile=/etc/supervisor.d/log/kafka_stdout.log
  stderr_logfile=/etc/supervisor.d/log/kafka_stderr.log
  
  

  创建日志目录

  mkdir -p /etc/supervisor.d/log
  

• 启动

  运行一下命令可以测试是否配置成功

  supervisord -c /etc/supervisord.conf
  

  之后可以查看要监控的进程是否已经启动

  supervisorctl
  

3. supervisord 命令行操作

• 启动supervisord进程

  supervisord -c /etc/supervisord.conf
  

• 查看进程状态

  supervisorctl
  

• 更多supervisorctl命令

  supervisorctl status
  supervisorctl stop all
  supervisorctl stop dbyw-python3
  supervisorctl start dbyw-python3
  supervisorctl restart dbyw-python3
  supervisorctl reload
  supervisorctl update
  

4. 设置supervisor 开启自启

vi /etc/rc.local

//添加
/usr/local/bin/supervisord -c /etc/supervisord.conf