Take the 2-minute tour ×
Server Fault is a question and answer site for system and network administrators. It's 100% free, no registration required.

I got this Flask app that I'm running with gunicorn on 127.0.0.1:5002 on this shared server. I got a nginx instance for myself that I can add custom configs to. The structure of the nginx directory is as such:

~/.nginx
    | conf.d
    | nginx.conf
    | [...]

Where nginx.conf contains the following:

pid       /media/sdl1/home/USERNAME/.nginx/pid;
error_log /media/sdl1/home/USERNAME/.nginx/error.log;

worker_processes     1;
worker_rlimit_nofile 65535;
events {
    worker_connections 8192;
}

http {
    include        /etc/nginx/mime.types;
    default_type   application/octet-stream;
    access_log     off;
    sendfile       on;
    tcp_nopush     on;
    real_ip_header X-Forwarded-For;

    client_max_body_size  1g;
    client_body_temp_path /media/sdl1/home/USERNAME/.nginx/client;
    fastcgi_temp_path     /media/sdl1/home/USERNAME/.nginx/fastcgi;
    uwsgi_temp_path       /media/sdl1/home/USERNAME/.nginx/uwscgi;
    scgi_temp_path        /media/sdl1/home/USERNAME/.nginx/scgi;

    include conf.d/*.conf;
}

Now that points to the conf.d directory which contains a file called 000-default-server.conf and an empty directory called 000-default-server.d

The 000-default-server.conf contains this:

server {
    listen      8080;
    server_name EXAMPLE.com;
    root        /media/sdl1/home/USERNAME/www/$host/public_html;
    index       index.html index.php;

    autoindex            on;
    autoindex_exact_size off;
    autoindex_localtime  on;
    error_log  /media/sdl1/home/USERNAME/.nginx/error.log  debug;

    # Pass files that end in .php to PHP
    location ~ \.php$ {
        fastcgi_read_timeout 1h;
        fastcgi_send_timeout 10m;

        include      /etc/nginx/fastcgi_params;
        fastcgi_pass unix:/media/sdl1/home/USERNAME/.nginx/php/socket;
    }

    location /apps/whatsmyuserinfo {
        proxy_redirect          off;
        proxy_pass_header       Server;
        proxy_set_header        Host            $http_host;
        proxy_set_header        X-Real-IP       $remote_addr;
        proxy_set_header        X-Scheme        $scheme;
        proxy_set_header        X-Forwarded-For $proxy_add_x_forwarded_for;
        client_max_body_size    10m;
        client_body_buffer_size 128k;
        proxy_connect_timeout   90;
        proxy_send_timeout      90;
        proxy_read_timeout      90;
        proxy_buffers           32 4k;
        proxy_pass              http://127.0.0.1:5002/;
    }

    # Deny access to anything starting with .ht
    location ~ /\.ht {
        deny  all;
    }

    include conf.d/000-default-server.d/*.conf;
}

With personal information replaced with less revealing values, those values are in all caps.

Now that's the general setup of nginx. Here's where my problem appears. If I use

curl -i 127.0.0.1:5002

through SSH on the server to see if gunicorn is properly serving the app, the app is properly served. If I do it externally, through EXAMPLE.com/apps/whatsmyuserinfo I get a 502 Bad Gateway error. The nginx error log could possibly shed some light on it, here is what it looks like for a single request to EXAMPLE.com/apps/whatsmyuserinfo on debug mode.

"GET / HTTP/1.0
Host: EXAMPLE.com
X-Real-IP: 10.0.0.1
X-Scheme: http
X-Forwarded-For: 188.176.255.124, 10.0.0.1
Connection: close
X-Host: EXAMPLE.com
X-Forwarded-Proto: http
User-Agent: Mozilla/5.0 (Windows NT 6.2; WOW64; rv:23.0) Gecko/20100101 Firefox/23.0
Accept: text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8
Accept-Language: en-US,en;q=0.5
Accept-Encoding: gzip, deflate
Cookie: __utma=18102207.768668361.1377199324.1377199324.1377199324.1; __utmz=18102207.1377199324.1.1.utmcsr=(direct)|utmccn=(direct)|utmcmd=(none)
Cache-Control: max-age=0

"
2013/08/23 19:00:01 [debug] 31195#0: *14 http cleanup add: 0000000001F44D30
2013/08/23 19:00:01 [debug] 31195#0: *14 get rr peer, try: 1
2013/08/23 19:00:01 [debug] 31195#0: *14 socket 10
2013/08/23 19:00:01 [debug] 31195#0: *14 epoll add connection: fd:10 ev:80000005
2013/08/23 19:00:01 [debug] 31195#0: *14 connect to 127.0.0.1:5002, fd:10 #15
2013/08/23 19:00:01 [debug] 31195#0: *14 http upstream connect: -2
2013/08/23 19:00:01 [debug] 31195#0: *14 posix_memalign: 0000000001F3D850:128 @16
2013/08/23 19:00:01 [debug] 31195#0: *14 event timer add: 10: 90000:1377284491613
2013/08/23 19:00:01 [debug] 31195#0: *14 http finalize request: -4, "/apps/whatsmyuserinfo?" a:1, c:2
2013/08/23 19:00:01 [debug] 31195#0: *14 http request count:2 blk:0
2013/08/23 19:00:01 [debug] 31195#0: *14 http run request: "/apps/whatsmyuserinfo?"
2013/08/23 19:00:01 [debug] 31195#0: *14 http upstream check client, write event:1, "/apps/whatsmyuserinfo"
2013/08/23 19:00:01 [debug] 31195#0: *14 http upstream recv(): -1 (11: Resource temporarily unavailable)
2013/08/23 19:00:01 [debug] 31195#0: *14 http upstream request: "/apps/whatsmyuserinfo?"
2013/08/23 19:00:01 [debug] 31195#0: *14 http upstream process header
2013/08/23 19:00:01 [error] 31195#0: *14 connect() failed (111: Connection refused) while connecting to upstream, client: 10.0.0.1, server: EXAMPLE.com, request: "GET /apps/whatsmyuserinfo HTTP/1.0", upstream: "http://127.0.0.1:5002/", host: "EXAMPLE.com"
2013/08/23 19:00:01 [debug] 31195#0: *14 http next upstream, 2
2013/08/23 19:00:01 [debug] 31195#0: *14 free rr peer 1 4
2013/08/23 19:00:01 [debug] 31195#0: *14 finalize http upstream request: 502
2013/08/23 19:00:01 [debug] 31195#0: *14 finalize http proxy request
2013/08/23 19:00:01 [debug] 31195#0: *14 free rr peer 0 0
2013/08/23 19:00:01 [debug] 31195#0: *14 close http upstream connection: 10
2013/08/23 19:00:01 [debug] 31195#0: *14 free: 0000000001F3D850, unused: 48
2013/08/23 19:00:01 [debug] 31195#0: *14 event timer del: 10: 1377284491613
2013/08/23 19:00:01 [debug] 31195#0: *14 reusable connection: 0
2013/08/23 19:00:01 [debug] 31195#0: *14 http finalize request: 502, "/apps/whatsmyuserinfo?" a:1, c:1
2013/08/23 19:00:01 [debug] 31195#0: *14 http special response: 502, "/apps/whatsmyuserinfo?"
2013/08/23 19:00:01 [debug] 31195#0: *14 http set discard body
2013/08/23 19:00:01 [debug] 31195#0: *14 xslt filter header
2013/08/23 19:00:01 [debug] 31195#0: *14 HTTP/1.1 502 Bad Gateway

Most relevant parts of that being:

2013/08/23 19:00:01 [debug] 31195#0: *14 http upstream recv(): -1 (11: Resource temporarily unavailable)
2013/08/23 19:00:01 [error] 31195#0: *14 connect() failed (111: Connection refused) while connecting to upstream, client: 10.0.0.1, server: EXAMPLE.com, request: "GET /apps/whatsmyuserinfo HTTP/1.0", upstream: "http://127.0.0.1:5002/", host: "EXAMPLE.com"
2013/08/23 19:00:01 [debug] 31195#0: *14 HTTP/1.1 502 Bad Gateway

Can anyone see what is happening here, because I don't understand why it refuses to work. I've tried making a new server in the nginx config specifically for the app and having the location be /, I've tried getting rid of most of the proxy settings and just leaving a few like proxy_pass, I've tried creating an upstream in the config and so on.

Full disclosure: I also posted this on stackoverflow, but was told to post on here aswell because it's more likely I would get help here.

share|improve this question

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.