2

I am trying to configure my django application on a new server. It works fine unless I try to pass GET parameters. I get the following errors.

uWSGI:

[pid: 21530|app: 0|req: 8/9] 109.68.173.7 () {42 vars in 880 bytes} [Thu Mar  2 17:19:29 2017] GET /install/?token=123&shop=1234&insales_id=124 => generated 0 bytes in 71 msecs (HTTP/1.1 500) 0 headers in 0 bytes (0 switches on core 0)

nginx:

2017/03/02 09:19:29 [error] 21644#0: *1 upstream prematurely closed connection while reading response header from upstream, client: 109.68.173.7, server: 151-248-112-157.xen.vps.regruhosting.ru, request: "GET /install/?token=123&shop=1234&insales_id=124 HTTP/1.1", upstream: "uwsgi://unix:/home/trackpost.sock:", host: "151-248-112-157.xen.vps.regruhosting.ru"

My config files. nginx.conf:

    # For more information on configuration, see:                                                                                               
    #   * Official English Documentation: http://nginx.org/en/docs/                                                                             
    #   * Official Russian Documentation: http://nginx.org/ru/docs/                                                                             

    user nginx;
    worker_processes auto;
    error_log /var/log/nginx/error.log;
    pid /var/run/nginx.pid;

    # Load dynamic modules. See /usr/share/nginx/README.dynamic.                                                                                
    include /usr/share/nginx/modules/*.conf;

    events {
        worker_connections  1024;
    }


    http {
        log_format  main  '$remote_addr - $remote_user [$time_local] "$request" '
                          '$status $body_bytes_sent "$http_referer" '
                          '"$http_user_agent" "$http_x_forwarded_for"';

        access_log  /var/log/nginx/access.log  main;

        sendfile            on;
        tcp_nopush          on;
        tcp_nodelay         on;
        keepalive_timeout   120;
        client_max_body_size 20M;

        uwsgi_read_timeout 86400;
        uwsgi_send_timeout 86400;

        proxy_buffers 8 32k;
        proxy_buffer_size 64k;

        include             /etc/nginx/mime.types;
        default_type        application/octet-stream;

        # Load modular configuration files from the /etc/nginx/conf.d directory.                                                                
        # See http://nginx.org/en/docs/ngx_core_module.html#include                                                                             
        # for more information.                                                                                                                 
        include /etc/nginx/conf.d/*.conf;

        server {
               listen 80;
               server_name 151-248-112-157.xen.vps.regruhosting.ru;
               location = favicon.ico { access_log off; log_not_found off; }
           location /static/ {
               root /home/rajansnow/django/trackpost;
           }
           location / {
              include uwsgi_params;
              uwsgi_pass unix:/home/trackpost.sock;
           }
    }
}

app.ini (uwsgi):

    [uwsgi]
project = trackpost
username = rajansnow
base = /home/%(username)/django

chdir = %(base)/%(project)
home = %(base)/venv
module = %(project).wsgi:application
plugin = python

master = true
processes = 5

uid = rajansnow
socket = /home/trackpost.sock
chown-socket = rajansnow:nginx
chmod-socket = 666
vacuum = true

touch-reload = /home/rajansnow/django/trackpost/uwsgi.ini
py-autoreload = 3
harakiri = 30

My code is a bit shitty, I know that. But it worked on my previous server, which was Debian based. The current is CentOS based. I've tried everything I found on SO, and no luck. What can I do to fix that?

  • Have you tried to just run `python manage.py runserver` and see, if Django handles GET requests properly within this particular virtual env? – abcdn Mar 02 '17 at 20:03
  • @abcdn hmmmm, that's an idea. I'll try it tomorrow –  Mar 02 '17 at 20:16
  • @abcdn strange thing: it doesn't work at all. Firefx says "Unable to connect" when accessing any page. The strangest is that it did work before I started configuring nginx and uwsgi. BTW, I stopped them, so that couldn't be a problem. No idea what's happening but I'll try to solve it –  Mar 03 '17 at 03:40
  • Try to bind it to any port other than 80 to make sure nothing is messed there, then debug step by step: django, uwsgi, nginx. Maybe some firewal settings? – abcdn Mar 03 '17 at 09:34
  • It took me a lot of time and nerves to debug it, @abcdn, the full story is in the answer. Thanks for reacting, man! –  Mar 05 '17 at 06:42

1 Answers1

1

It's time for mysteries, kids!

The problem mentioned in the question was solved by hard-checking every project file. It turned out that some of them contained unexpected symbols, which used to make Django throw encoding errors, and, as a result, make uwsgi stop.

The mystery part is that the files had been last modified a long time before the problem occured. So how could it happen that they contain extra symbols? Nobody knows.

At least for now, we can say it's solved.

  • For such cases it is good to have Django Logging turned on, as with `DEBUG=False`, which is normally the case on production, it doesn't show any errors: https://docs.djangoproject.com/en/1.10/topics/logging/#examples – abcdn Mar 05 '17 at 09:58
  • @abcdn it was turned on, and that's a bit weired –  Mar 05 '17 at 10:12