[Answer]-What is the best "architecture" for running celery (and celery beat) in production environment (debian server)?

by

0đź‘Ť

âś…

Based on Chris’s answer, the solution how to restart supervisord program without sudo permission:
You have to edit supervisord.conf to change socket permissions (in my case located at /etc/supervisor/supervisord.conf]

[unix_http_server]
file=/var/run//supervisor.sock   ; (the path to the socket file)
chmod=0766                       ; sockef file mode (default 0700)

Then you have to make sure, that user who is written in config file is user who will restart program:

[program:projectx_celerybeat]
; Set full path to celery program if using virtualenv
command=/path_to_project_root/env/bin/celery beat -A main -s /path_to_project_root/celerybeat-schedule --loglevel=INFO

; remove the -A myapp argument if you are not using an app instance

directory=/home/xxx/project_root/celery_root/
user=YOUR_USER
numprocs=1
stdout_logfile=/path_to_log/beat.log
stderr_logfile=/path_to_log/beat.log
autostart=true
autorestart=true
startsecs=10

Then this command runned by YOUR_USER shoud work:

supervisorctl [stop/start/restart] [program_name]
👤snakey

1đź‘Ť

I haven’t tried this specifically with celery beat, but I seem to recall it worked for me on another program. You can configure supervisor to run the program under a particular user and then I believe you can use the following command without requiring sudo if it is run from that user:
supervisorctl [stop/start/restart] [program_name]. See the “user” parameter here:

http://supervisord.org/configuration.html#program-x-section-settings

👤Chris Ward

Leave a comment