【已采納】supervisor在服務器端（linux），如何一直運行你的python代碼

一、安裝

• 源碼安裝

  先下載最新的supervisor安裝包：https://pypi.python.org/pypi/supervisor 

　　 如：

　　（python3命令為 pip install git+https://github.com/Supervisor/supervisor@master）或者pip install supervisor(pip2.7版本可用)

1.  
   cd /usr/local/ENV
2.  
   wget https://pypi.python.org/packages/7b/17/88adf8cb25f80e2bc0d18e094fcd7ab300632ea00b601cbbbb84c2419eae/supervisor-3.3.2.tar.gz
3.  
   tar -zxvf supervisor-3.3.2.tar.gz
4.  
   cd supervisor-3.3.2
5.  
   python setup.py install  #本地python版本為python2.7
6.  
   # python2.7 setup.py install #本地python版本為python3以上

二、配置

1.生成配置文件

　echo_supervisord_conf > /etc/supervisord.conf

2.啟動

supervisord -c /etc/supervisord.conf（這一步可以放在3.配置文件之后噢，可以先更改配置文件，再啟動） 

 

查看 supervisord 是否在運行：

ps aux | grep supervisord 如下圖即可：

 

 

 

3.配置

打開配置文件

vim /etc/supervisord.conf

 

在最下邊加入：舉例子我的項目名稱為：love.py 目錄文件名為：lowerlove

[program:lowerlove]   #lowerlove 為程序的名稱
command=python /usr/local/ENV/lowerlove/love.py  #需要執行的命令
directory=/usr/local/ENV/lowerlove  #命令執行的目錄
environment=ASPNETCORE__ENVIRONMENT=Production #環境變量
user=root 　　#用戶
stopsignal=10 #這個是當我們向子進程發送stopsignal信號后，到系統返回信息給supervisord，所等待的最大時間。 超過這個時間，supervisord會向該子進程發送一個強制kill的信號。根據自己項目性能情況實際修改

autostart=true #是否自啟動c　　

autorestart=true #是否自動重啟

startsecs=3 #自動重啟時間間隔（s）
stderr_logfile=/usr/local/ENV/lowerlove/love.err.log #錯誤日志文件
stdout_logfile=/usr/local/ENV/lowerlove/love.out.log #輸出日志文件

如報錯：

 unlink /tmp/supervisor.sock

在配置文件底部，配置include

1.  
   [include]
2.  
   files=/etc/supervisor/*.conf  #若你本地無/etc/supervisor目錄，請自建

用supervisor管理進程，配置如下：

1.  
   cd /etc/supervisor
2.  
   vim ossfs.conf  # 這里的文件名稱自定義

加入以下內容：

1.  
   ; 設置進程的名稱，使用 supervisorctl 來管理進程時需要使用該進程名
2.  
   [program:your_program_name]
3.  
   command=python server.py --port=9000
4.  
   ;numprocs=1 ; 默認為1
5.  
   ;process_name=%(program_name)s ; 默認為 %(program_name)s，即 [program:x] 中的 x
6.  
   directory=/home/python/tornado_server ; 執行  command 之前，先切換到工作目錄
7.  
   user=oxygen ; 使用 oxygen 用戶來啟動該進程
8.  
   ; 程序崩潰時自動重啟，重啟次數是有限制的，默認為3次
9.  
   autorestart= true
10.  
    redirect_stderr= true ; 重定向輸出的日志
11.  
    stdout_logfile = /var/ log/supervisord/tornado_server.log
12.  
    loglevel=info

這里是啟動要配置的參數，請根據自己的項目自定義添加

更改了supervisor配置文件，需要重啟,運行以下指令：

supervisorctl reload 

 

4.supervisorctl的用法（這個是重點，熟練使用必須記住）

1.  
   supervisord : 啟動supervisor
2.  
   supervisorctl reload :修改完配置文件后重新啟動supervisor
3.  
   supervisorctl status :查看supervisor監管的進程狀態
4.  
   supervisorctl  start 進程名 ：啟動XXX進程
5.  
   supervisorctl  stop 進程名 ：停止XXX進程
6.  
   supervisorctl  stop all：停止全部進程，注：start、restart、stop都不會載入最新的配置文件。
7.  
   supervisorctl  update：根據最新的配置文件，啟動新配置或有改動的進程，配置沒有改動的進程不會受影響而重啟

5.若不使用控制台來管理進程，用瀏覽器來管理，該如何配置？

打開配置文件

vim /etc/supervisord.conf 

 

配置 inet_http_server

1.  
   [ inet_http_server]
2.  
   port= 127.0.0.1:9001 ; 服務器ip
3.  
   username=xxx ;自定義
4.  
   password=xxx ;自定義

三、設置開機啟動

vim /etc/init.d/supervisord 

 

添加以下腳本

1.  
   #! /bin/sh
2.  
   ### BEGIN INIT INFO
3.  
   # Provides: supervisord
4.  
   # Required-Start: $remote_fs
5.  
   # Required-Stop: $remote_fs
6.  
   # Default-Start: 2 3 4 5
7.  
   # Default-Stop: 0 1 6
8.  
   # Short-Description: Example initscript
9.  
   # Description: This file should be used to construct scripts to be
10.  
    # placed in /etc/init.d.
11.  
    ### END INIT INFO
12.  
     
13.  
    # Author: Dan MacKinlay <danielm@phm.gov.au>
14.  
    # Based on instructions by Bertrand Mathieu
15.  
    # http://zebert.blogspot.com/2009/05/installing-django-solr-varnish-and.html
16.  
     
17.  
    # Do NOT "set -e"
18.  
     
19.  
    # PATH should only include /usr/* if it runs after the mountnfs.sh script
20.  
    PATH=/usr/ local/sbin:/usr/local/bin:/sbin:/usr/sbin:/bin:/usr/bin
21.  
    DESC= "Description of the service"
22.  
    NAME=supervisord
23.  
    DAEMON=/usr/ local/bin/supervisord
24.  
    DAEMON_ARGS= " -c /etc/supervisord.conf"
25.  
    #PIDFILE=/var/run/$NAME.pid
26.  
    PIDFILE=/tmp/ $NAME.pid
27.  
    SCRIPTNAME=/etc/init.d/ $NAME
28.  
     
29.  
    # Exit if the package is not installed
30.  
    [ -x  "$DAEMON" ] || exit 0
31.  
     
32.  
    # Read configuration variable file if it is present
33.  
    [ -r /etc/default/ $NAME ] && . /etc/default/$NAME
34.  
     
35.  
    # Load the VERBOSE setting and other rcS variables
36.  
    . /lib/init/vars.sh
37.  
     
38.  
    # Define LSB log_* functions.
39.  
    # Depend on lsb-base (>= 3.0-6) to ensure that this file is present.
40.  
    . /lib/lsb/init-functions
41.  
     
42.  
    #
43.  
    # Function that starts the daemon/service
44.  
    #
45.  
    do_start()
46.  
    {
47.  
    # Return
48.  
    # 0 if daemon has been started
49.  
    # 1 if daemon was already running
50.  
    # 2 if daemon could not be started
51.  
    start-stop-daemon --start --quiet --pidfile  $PIDFILE --exec $DAEMON --test > /dev/null \
52.  
    ||  return 1
53.  
    start-stop-daemon --start --quiet --pidfile  $PIDFILE --exec $DAEMON -- \
54.  
    $DAEMON_ARGS \
55.  
    ||  return 2
56.  
    # Add code here, if necessary, that waits for the process to be ready
57.  
    # to handle requests from services started subsequently which depend
58.  
    # on this one. As a last resort, sleep for some time.
59.  
    }
60.  
     
61.  
    #
62.  
    # Function that stops the daemon/service
63.  
    #
64.  
    do_stop()
65.  
    {
66.  
    # Return
67.  
    # 0 if daemon has been stopped
68.  
    # 1 if daemon was already stopped
69.  
    # 2 if daemon could not be stopped
70.  
    # other if a failure occurred
71.  
    start-stop-daemon --stop --quiet --retry=TERM/30/KILL/5 --pidfile  $PIDFILE --name $NAME
72.  
    RETVAL= "$?"
73.  
    [  "$RETVAL" = 2 ] && return 2
74.  
    # Wait for children to finish too if this is a daemon that forks
75.  
    # and if the daemon is only ever run from this initscript.
76.  
    # If the above conditions are not satisfied then add some other code
77.  
    # that waits for the process to drop all resources that could be
78.  
    # needed by services started subsequently. A last resort is to
79.  
    # sleep for some time.
80.  
    start-stop-daemon --stop --quiet --oknodo --retry=0/30/KILL/5 -- exec $DAEMON
81.  
    [  "$?" = 2 ] && return 2
82.  
    # Many daemons don't delete their pidfiles when they exit.
83.  
    rm -f  $PIDFILE
84.  
    return "$RETVAL"
85.  
    }
86.  
     
87.  
    #
88.  
    # Function that sends a SIGHUP to the daemon/service
89.  
    #
90.  
    do_reload() {
91.  
    #
92.  
    # If the daemon can reload its configuration without
93.  
    # restarting (for example, when it is sent a SIGHUP),
94.  
    # then implement that here.
95.  
    #
96.  
    start-stop-daemon --stop --signal 1 --quiet --pidfile  $PIDFILE --name $NAME
97.  
    return 0
98.  
    }
99.  
     
100.  
     case "$1" in
101.  
     start)
102.  
     [  "$VERBOSE" != no ] && log_daemon_msg "Starting $DESC" "$NAME"
103.  
     do_start
104.  
     case "$?" in
105.  
     0|1) [  "$VERBOSE" != no ] && log_end_msg 0 ;;
106.  
     2) [  "$VERBOSE" != no ] && log_end_msg 1 ;;
107.  
     esac
108.  
     ;;
109.  
     stop)
110.  
     [  "$VERBOSE" != no ] && log_daemon_msg "Stopping $DESC" "$NAME"
111.  
     do_stop
112.  
     case "$?" in
113.  
     0|1) [  "$VERBOSE" != no ] && log_end_msg 0 ;;
114.  
     2) [  "$VERBOSE" != no ] && log_end_msg 1 ;;
115.  
     esac
116.  
     ;;
117.  
     #reload|force-reload)
118.  
     #
119.  
     # If do_reload() is not implemented then leave this commented out
120.  
     # and leave 'force-reload' as an alias for 'restart'.
121.  
     #
122.  
     #log_daemon_msg "Reloading $DESC" "$NAME"
123.  
     #do_reload
124.  
     #log_end_msg $?
125.  
     #;;
126.  
     restart|force-reload)
127.  
     #
128.  
     # If the "reload" option is implemented then remove the
129.  
     # 'force-reload' alias
130.  
     #
131.  
     log_daemon_msg  "Restarting $DESC" "$NAME"
132.  
     do_stop
133.  
     case "$?" in
134.  
     0|1)
135.  
     do_start
136.  
     case "$?" in
137.  
     0) log_end_msg 0 ;;
138.  
     1) log_end_msg 1 ;;  # Old process is still running
139.  
     *) log_end_msg 1 ;;  # Failed to start
140.  
     esac
141.  
     ;;
142.  
     *)
143.  
     # Failed to stop
144.  
     log_end_msg 1
145.  
     ;;
146.  
     esac
147.  
     ;;
148.  
     *)
149.  
     #echo "Usage: $SCRIPTNAME {start|stop|restart|reload|force-reload}" >&2
150.  
     echo "Usage: $SCRIPTNAME {start|stop|restart|force-reload}" >&2
151.  
     exit 3
152.  
     ;;
153.  
     esac
154.  
      
155.  
     :

1.  
   # 設置該腳本為可以執行
2.  
   sudo chmod +x /etc/init.d/supervisord
3.  
   # 設置為開機自動運行
4.  
   sudo update-rc.d supervisord defaults
5.  
   # 試一下，是否工作正常
6.  
   service supervisord stop
7.  
   service supervisord start

若報錯：insserv: warning: script 'service' missing LSB tags and overrides，請執行:

sudo apt-get remove insserv

 

 

更多supervisord配置可以參考 https://www.cnblogs.com/ajianbeyourself/p/5534737.html

 

 

 

聲明：本文為博主學習感悟總結，水平有限，如果不當，歡迎指正。如果您認為還不錯，歡迎轉載。轉載與引用請注明作者及出處。