Client receives 3001

[boris-savic]

We're facing an issue on our deployment servers that we just can't wrap our heads around.

The server.py is bein run via supervisor using following config:

[program:swamp_dragon]
directory=/home/olaii/pay_backend/releases/current/pay_backend/
command=/home/olaii/pay_backend/env/bin/python /home/olaii/pay_backend/releases/current/pay_backend/manage.py runsd

user=olaii
autostart=true
autorestart=true

stdout_logfile=/var/log/supervisor/pay_backend_swampdragon_stdout.log
stderr_logfile=/var/log/supervisor/pay_backend_swampdragon_stderr.log

environment=HOME='/home/olaii/pay_backend/releases/current/pay_backend/'
process_name=%(program_name)s

This all seems fine, except we're not seeing anything in stdout_logfile nor in stderr_logfile - they're just empty.

At first our clients can connect to the socket without a problem, but after a while all of the clients start getting 3001 Connection aborted on subscribe (we're testing with 3-4 users). This keeps happening all the time and only after we restart swampdragon supervisor process thigs go back to normal - for about 30min to 1h and then it's down again....

What could be the problem here? I don't beleive that restarting the process every now and then is a solution here...

Thanks and keep up the good work!

[boris-savic]

Found the reason for this. It's loosing MySQL connection as described in https://github.com/jonashagstedt/swampdragon/issues/2

Has this been solved in any other way or should I reffer to the http://swampdragon.net/blog/mysql-server-has-gone-away/ and fix it with this?