Gunicorn internal server error

Hello, guys. Don’t know what’s wrong.
I am using default digital ocean setup. I’ve used it a lot of times by now. But this time I needed to implement post request from my client mobile app. I’ve installed cors-headers app for django and tested my app locally for a long time, everything was great (btw my mobile app is written on ionic framework and uses angularjs to post). On server everything was also ok with following commands:
python manage.py runserver 9000
and (!!!)
gunicorn config.wsgi —bind 127.0.0.1:9000
By ok I mean post requests were handled as they should be.
But when I run gunicorn as a service (sudo service gunicorn restart) on every post request from my mobile app I get 500 internal server error. And this is just after everything was ok not in service mode. What’s so different in service mode so it doesn’t allow my post requests to reach server?
Again: the difference is only service / not service mode of gunicorn. (but of cource service mode has it’s config while «working» gunicorn is just «gunicorn config.wsgi —bind 127.0.0.1:9000»)

Also I need to say that the website is running fine even in service mode. So the problem is only with this cross-domain post requests.

Errors examples (gunicorn as a service):

178.140.183.40 - - [29/Jun/2015:22:07:10 -0400] "POST /create_o_action/ HTTP/1.1" 500 58581 "-" "Mozilla/5.0 (Linux; Android 5.1.1; Nexus 5 Build/LMY48B; wv) AppleWebKit/537.36 (KHTML, like Gecko) Version/4.0 Chrome/43.0.2357.121 Mobile Safari/537.36"
178.140.183.40 - - [29/Jun/2015:22:07:10 -0400] "POST /create_o_action/ HTTP/1.1" 500 58595 "-" "Mozilla/5.0 (Linux; Android 5.1.1; Nexus 5 Build/LMY48B; wv) AppleWebKit/537.36 (KHTML, like Gecko) Version/4.0 Chrome/43.0.2357.121 Mobile Safari/537.36"

Success examples (gunicorn config.wsgi —bind 127.0.0.1:9000 ):

178.140.183.40 - - [29/Jun/2015:22:25:58 -0400] "POST /create_o_action/ HTTP/1.1" 200 32 "-" "Mozilla/5.0 (Linux; Android 5.1.1; Nexus 5 Build/LMY48B; wv) AppleWebKit/537.36 (KHTML, like Gecko) Version/4.0 Chrome/43.0.2357.121 Mobile Safari/537.36"
178.140.183.40 - - [29/Jun/2015:22:25:58 -0400] "POST /create_c_action/ HTTP/1.1" 200 34 "-" "Mozilla/5.0 (Linux; Android 5.1.1; Nexus 5 Build/LMY48B; wv) AppleWebKit/537.36 (KHTML, like Gecko) Version/4.0 Chrome/43.0.2357.121 Mobile Safari/537.36"

My gunicorn settings (more or less default digital ocean):

/etc/init/gunicorn.conf

description "Gunicorn daemon for Django project"

start on (local-filesystems and net-device-up IFACE=eth0)
stop on runlevel [!12345]

# If the process quits unexpectadly trigger a respawn
respawn

setuid django
setgid django
chdir /home/django/counter_platform_django

exec gunicorn 
    --name=webapp 
    --pythonpath=webapp 
    --bind=127.0.0.1:9000 
    --config /etc/gunicorn.d/gunicorn.py 
    config.wsgi:application

/etc/gunicorn.d/gunicorn.py

"""gunicorn WSGI server configuration."""
from multiprocessing import cpu_count
from os import environ


def max_workers():
    return cpu_count() * 2 + 1

max_requests = 1000
worker_class = 'gevent'
workers = max_workers()

So what’s exactly different in two ways of running gunicorn? I’m really curious. Of course there is —daemon (which btw also works fine), but solution will be better.
Also I found a lot of questions on the internet (all without answers) about problems with post requests and gunicorn in service mode. If we will find solution — some of them also can be answered.

Это мое первое развертывание в Django для производства.

Я проверил, может ли он обслуживать мое приложение Django, выполнив следующую команду, и нет никаких проблем, я могу добраться до моего сайта по ip-адресу моего сервера.

gunicorn hello.wsgi:application --bind 0.0.0.0:8000

application --bind 0.0.0.0:8000
[2022-05-01 11:00:46 +0000] [26484] [INFO] Starting gunicorn 20.1.0
[2022-05-01 11:00:46 +0000] [26484] [INFO] Listening at: http://0.0.0.0:8000 (26484)
[2022-05-01 11:00:46 +0000] [26484] [INFO] Using worker: sync
[2022-05-01 11:00:46 +0000] [26486] [INFO] Booting worker with pid: 26486
/home/scknylmz35/PersonelBlog/home/views.py:24: UnorderedObjectListWarning: Pagination may yield inconsistent results with an unordered object_list: <class 'home.models.Blog'> QuerySet.

Затем я выполнил следующие команды:

sudo nano /etc/systemd/system/gunicorn.service

[Unit]
Description=gunicorn daemon
After=network.target

[Service]
User=baris
Group=www-data
WorkingDirectory=/home/baris/eventhub
ExecStart=/home/baris/eventhub/venv/bin/gunicorn --access-logfile - --workers 3 --bind unix:/home/baris/eventhub/eventhub.sock eventhub.wsgi:application

[Install]
WantedBy=multi-user.target

sudo systemctl start gunicorn
sudo systemctl enable gunicorn

И я протестировал gunicorn

>> sudo systemctl status gunicorn

 Loaded: loaded (/etc/systemd/system/gunicorn.service; enabled; vendor pres>
     Active: active (running) since Sun 2022-05-01 08:01:47 UTC; 2h 19min ago
   Main PID: 23885 (gunicorn)
      Tasks: 4 (limit: 1132)
     Memory: 136.3M
     CGroup: /system.slice/gunicorn.service
             ├─23885 /home/scknylmz35/PersonelBlog/venv/bin/python /home/scknyl>
             ├─23897 /home/scknylmz35/PersonelBlog/venv/bin/python /home/scknyl>
             ├─23898 /home/scknylmz35/PersonelBlog/venv/bin/python /home/scknyl>
             └─23899 /home/scknylmz35/PersonelBlog/venv/bin/python /home/scknyl>

Кажется, что все работает хорошо. Но когда я пытаюсь зайти на свой сайт как mysite.me, я сталкиваюсь с «Internal Server Error»

Также, когда я проверяю журнал ошибок nginx, ошибки нет.

Как я могу найти, в чем проблема? Не могли бы вы направить меня?

Вернуться на верх