Question
Gunicorn unable to start after server update ( 502 bad gateway )
Hello, I have updated my old Ubuntu server from version 14 to 16. This also included a Django and Python update. After restarting the server, I got a 502 error from Nginx, which I have discovered is being caused by Gunicorn not being able to start. I am able to start gunicorn manually, and the site works perfectly, by doing:
root@server:/home/django/django_project/django_project# gunicorn -b 127.0.0.1:9000 wsgi
wsgi.py is located at /home/django/django_project/django_project
The problem I am having is that the gunicorn service is not able to start automatically, even if the config (/etc/init/gunicorn.conf) looks correct:
[...]
chdir /home/django
exec gunicorn \
--name=django_project \
--pythonpath=django_project \
--bind=127.0.0.1:9000 \
--config /etc/gunicorn.d/gunicorn.py \
django_project.wsgi:application
Any ideas of why the service is not starting by itself anymore?
service gunicorn restart
does not work either.
These answers are provided by our Community. If you find them useful, show some love by clicking the heart. If you run into issues leave a comment, or add your own answer to help others.
So, just for everyone to know, I finally solved this by editing /etc/init/gunicorn.conf
The file was assuming gunicorn was in my django folder, but in reality it is located at /usr/bin
So, the old file was:
setuid django
setgid django
chdir /home/django
exec gunicorn \
--name=django_project \
--pythonpath=django_project \
--bind=127.0.0.1:9000 \
--config /etc/gunicorn.d/gunicorn.py \
django_project.wsgi:application
The new file is now:
setuid django
setgid django
chdir /home/django/django_project
exec usr/bin/gunicorn \
--bind=127.0.0.1:9000 \
--config /etc/gunicorn.d/gunicorn.py \
django_project.wsgi:application
So I made it execute gunicorn from its real location ( it was not there before the update ) and I also cleaned up the command by removing name and pythonpath.
Hello, first of all, the gunicorn log file is not getting written anymore:
/var/log/upstart/gunicorn.log
Regarding the log files you mentioned ( journalctl -xe ), what I am getting is:
- Unit gunicorn.service has begun starting up.
Nov 66 06:33:09 PythonServer gunicorn[9300]: * Starting Gunicorn workers
Nov 66 06:33:09 PythonServer gunicorn[9300]: vetc’gunicorn.d/gunicorn.py:2: RuntimeWarning: Parent nodule ’ etc/gunicorn.d/gunic
Nov 66 06:33:09 PythonServer gunicorn[9300]: from multiprocessing import cpu_count
Nov 06 06:33:09 PythonServer gunicorn(9300]: vetc/gunicorn.d/gunicorn.py:3: RuntimeWarning: Parent module ‘etc’gunicorn.d/gunic
Nov 66 06:33:09 PythonServer gunicorn[9300]: fron os import environ
Nov 66 06:33:09 PythonServer gunicorn[9300]:
Nov 06 06:33:09 PythonServer systend(1]: Started gunicorn.service.
— Subject: Unit gunicorn.service has finished start-up
- Def ined-By: systend
- Support: http:// lists .freedesktop.org/mailman/ list info/systemd—devel
- Unit gunicorn.service has finished starting up.
- The start-up result is done.
Nov 66 06:33:46 PythonServer sshd(93161]: pam_unix(sshd:auth): authentication failure: logname= uid=8 euid=0 tty=ssh ruser= rhost
Nov 66 06:33:43 PythonServer sshd[93141]: pam_unix(sshd:auth): authentication failure: logname= uid=@ euid=0 tty=ssh ruser= rhost
Nov 66 06:33:43 PythonServer sshd(93161]: Failed password for root fron 51.38.186.207 port 37778 ssh2
Nov 66 06:33:43 PythonServer sshd(9316): Received disconnect fron 51.38.186.267 port 37778:11: Bye Bye Cpreauth]
Nov 06 06:33:43 PythonServer sshd(93161]: Disconnected from 51.38.186.207 port 37778 Cpreauth]
Nov 06 06:33:44 PythonServer sshd[9314]: Failed password for root from 210.177.54.141 port 50740 sshz
Nov 66 06:33:44 PythonServer sshd[9314]1: Received disconnect from 210.177.54.141 port 50740:11: Bye Bye [preauth]
Nov 06 06:33:44 PythonServer sshd([9314]: Disconnected from 216.177.54.141 port 56740 [preauth]
lines 989 1035/1035 CEND)
I also had another idea: Maybe the django user had no permission to run gunicorn ( because I am manually running it as root ), but when I do su django and run the manual gunicorn command, it still runs OK.
Hi @baroale,
Can you show us the error you are experiencing when you try to perform
service gunicorn restart
If it doesn’t start, it should at least have an error being logged somewhere, perhaps
journalctl -xe
Regards, KDSys
Edit: I meant wsgi.py is located at /home/django/django_project/django_project