# TCP Connection **URL:** \ **Category:** NGINX\ **Created:** [March 4, 2016, 5:21pm UTC](https://www.ruby-forum.com/t/tcp-connection/244203 "2016-03-04T17:21:02Z")\ **Posts on this page:** 11\ **Page:** 1
**Author:** ![Sadaf\_N](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/s/858c86/32.png) [@Sadaf\_N](https://www.ruby-forum.com/u/Sadaf_N)\ **Post date:** [March 4, 2016, 5:21pm UTC](https://www.ruby-forum.com/t/tcp-connection/244203/1 "2016-03-04T17:21:02Z")
Hi All, If I’m sending this email to wrong list, apologies and give me the good one. I have one embedded application which needs to connect via TCP and send an HTTP request. We run Nginx on Ubuntu 14.04. I can connect and persist, but once the packet is sent, I get HTTP 400 without any log and without web server taking the request. I think it is a deeper level on Nginx but having hard to find a good answer. Hope somebody can help. Thanks Giulio ---
**Author:** ![Giulio\_L](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/g/0ea827/32.png) [@Giulio\_L](https://www.ruby-forum.com/u/Giulio_L)\ **Post date:** [March 4, 2016, 5:25pm UTC](https://www.ruby-forum.com/t/tcp-connection/244203/2 "2016-03-04T17:25:43Z")
On Friday 04 March 2016 17:20:18 Giulio L. wrote: > Hope somebody can help. [..] The reason of 400 is usually logged on “info” level in error\_log. See: [Core functionality](http://nginx.org/r/error_log) wbr, Valentin V. Bartenev ---
**Author:** ![Giulio\_L](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/g/0ea827/32.png) [@Giulio\_L](https://www.ruby-forum.com/u/Giulio_L)\ **Post date:** [March 4, 2016, 5:42pm UTC](https://www.ruby-forum.com/t/tcp-connection/244203/3 "2016-03-04T17:42:04Z")
Thank you for your answer. Here is my log after the request, my request, my response and my nginx.conf root@vps190138:~# cat /var/log/nginx-error.log 2016/03/04 15:49:02 [notice] 8990#0: using the “epoll” event method 2016/03/04 15:49:02 [notice] 8990#0: nginx/1.8.1 2016/03/04 15:49:02 [notice] 8990#0: OS: Linux 3.13.0-66-generic 2016/03/04 15:49:02 [notice] 8990#0: getrlimit(RLIMIT\_NOFILE): 1024:4096 2016/03/04 15:49:02 [notice] 8991#0: start worker processes 2016/03/04 15:49:02 [notice] 8991#0: start worker process 8993 #REQUEST AT+CIPSTART=“TCP”, “[myhost.com](http://myhost.com)”,“80” AT+CIPSEND=150 POST /devices/deviceRegister HTTP/1.1 Host: [myhost.com](http://myhost.com) Cache-Control: no-cache Content-Type: application/json Content-Length:14 {“IMEI”:“aa”} #RESPONSE SEND OK HTTP/1.1 400 Bad Request Server: nginx/1.8.1 Date: Fri, 04 Mar 2016 15:52:21 GMT Content-Type: text/html Content-Length: 172 Connection: close 400 Bad Request # 400 Bad Request * * * nginx/1.8.1 ---
**Author:** ![Giulio\_L](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/g/0ea827/32.png) [@Giulio\_L](https://www.ruby-forum.com/u/Giulio_L)\ **Post date:** [March 4, 2016, 6:56pm UTC](https://www.ruby-forum.com/t/tcp-connection/244203/4 "2016-03-04T18:56:51Z")
On Friday 04 March 2016 17:41:42 Giulio L. wrote: > 2016/03/04 15:49:02 [notice] 8991#0: start worker process 8993 [..] You have another error\_log directive on the “http” level of your configuration, and it overrides log settings set on the main config level. wbr, Valentin V. Bartenev ---
**Author:** ![Giulio\_L](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/g/0ea827/32.png) [@Giulio\_L](https://www.ruby-forum.com/u/Giulio_L)\ **Post date:** [March 4, 2016, 9:48pm UTC](https://www.ruby-forum.com/t/tcp-connection/244203/5 "2016-03-04T21:48:00Z")
## You have been told to use the info error\_log level, which is not set (defaults to error). ​Make sure no other error\_log directive might overwrite this one.​ _B. R._ On Fri, Mar 4, 2016 at 7:12 PM, Giulio L. [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) ---
**Author:** ![Giulio\_L](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/g/0ea827/32.png) [@Giulio\_L](https://www.ruby-forum.com/u/Giulio_L)\ **Post date:** [March 4, 2016, 7:13pm UTC](https://www.ruby-forum.com/t/tcp-connection/244203/6 "2016-03-04T19:13:12Z")
Still having no luck. No log is generated. TCP connection has any relation to web socket support configuration ? this is my site config. server { listen 80; server\_name [myhost.com](http://myhost.com); access\_log /var/log/nginx/myhost.com.log combined; error\_log /var/log/nginx/myhost.com.error.log; ``` root /home/spark/myhost.com/web; try_files $uri /index.php; set $cache_uri $request_uri; if ($request_method = POST) { set $cache_uri 'null cache'; } location ~ \.php$ { try_files $uri =404; fastcgi_split_path_info ^(.+\.php)(/.+)$; # With php5-fpm: fastcgi_pass unix:/var/run/php5-fpm.sock; fastcgi_index index.php; include fastcgi_params; fastcgi_param SCRIPT_FILENAME ``` $document\_root$fastcgi\_script\_name; ``` # WebSocket support — not working #proxy_pass http://127.0.0.1:8765; #proxy_redirect off; #proxy_http_version 1.1; #proxy_set_header Upgrade $http_upgrade; #proxy_set_header Connection "upgrade"; #proxy_buffering off; #debug requests #echo_duplicate 1 $echo_client_request_headers; #echo "\r"; #echo_read_request_body; #echo $request_body; } ``` } ---
**Author:** ![Giulio\_L](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/g/0ea827/32.png) [@Giulio\_L](https://www.ruby-forum.com/u/Giulio_L)\ **Post date:** [March 4, 2016, 11:24pm UTC](https://www.ruby-forum.com/t/tcp-connection/244203/7 "2016-03-04T23:24:59Z")
On Friday 04 March 2016 19:12:56 Giulio L. wrote: > ``` > access_log /var/log/nginx/myhost.com.log combined; > error_log /var/log/nginx/myhost.com.error.log; > > ``` [..] You have another error\_log directive on the “server” level that overrides the directives from previous levels. wbr, Valentin V. Bartenev ---
**Author:** ![Giulio\_L](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/g/0ea827/32.png) [@Giulio\_L](https://www.ruby-forum.com/u/Giulio_L)\ **Post date:** [March 4, 2016, 11:26pm UTC](https://www.ruby-forum.com/t/tcp-connection/244203/8 "2016-03-04T23:26:09Z")
Here is my nginx.conf and site config. After calling tcp connection and getting http 400, no log is generated. I’m following Thanks server { listen 80; server\_name apidvc.sparkgo.cc; access\_log /var/log/nginx/myhost.com.access.log combined; ``` root /home/spark/myhost.com/web; try_files $uri /index.php; set $cache_uri $request_uri; if ($request_method = POST) { set $cache_uri 'null cache'; } location ~ \.php$ { try_files $uri =404; fastcgi_split_path_info ^(.+\.php)(/.+)$; # NOTE: You should have "cgi.fix_pathinfo = 0;" in ``` php.ini ``` # With php5-fpm: fastcgi_pass unix:/var/run/php5-fpm.sock; fastcgi_index index.php; include fastcgi_params; fastcgi_param SCRIPT_FILENAME ``` $document\_root$fastcgi\_script\_name; ``` #debug request #echo_duplicate 1 $echo_client_request_headers; #echo "\r"; #echo_read_request_body; #echo $request_body; } ``` } The file exists and it is fed as you can see below when I restart nginx: root@vps190138:~# cat /var/log/nginx-error.log 2016/03/04 21:14:55 [notice] 13900#0: signal 3 (SIGQUIT) received, shutting down 2016/03/04 21:14:55 [notice] 13902#0: gracefully shutting down 2016/03/04 21:14:55 [notice] 13902#0: exiting 2016/03/04 21:14:55 [notice] 13902#0: exit 2016/03/04 21:14:55 [notice] 13900#0: signal 15 (SIGTERM) received, exiting 2016/03/04 21:14:55 [notice] 13902#0: signal 15 (SIGTERM) received, exiting 2016/03/04 21:14:55 [notice] 13900#0: signal 17 (SIGCHLD) received 2016/03/04 21:14:55 [notice] 13900#0: worker process 13902 exited with code 0 2016/03/04 21:14:55 [notice] 13900#0: exit 2016/03/04 21:14:55 [notice] 13949#0: using the “epoll” event method 2016/03/04 21:14:55 [notice] 13949#0: nginx/1.8.1 2016/03/04 21:14:55 [notice] 13949#0: OS: Linux 3.13.0-66-generic 2016/03/04 21:14:55 [notice] 13949#0: getrlimit(RLIMIT\_NOFILE): 1024:4096 2016/03/04 21:14:55 [notice] 13950#0: start worker processes 2016/03/04 21:14:55 [notice] 13950#0: start worker process 13952 root@vps190138:~# ---
**Author:** ![Giulio\_L](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/g/0ea827/32.png) [@Giulio\_L](https://www.ruby-forum.com/u/Giulio_L)\ **Post date:** [March 4, 2016, 10:49pm UTC](https://www.ruby-forum.com/t/tcp-connection/244203/9 "2016-03-04T22:49:20Z")
does /var/log/nginx exist, and it is writeable by the web server user? On 03/05/2016 07:12 AM, Giulio L. wrote: > ``` > access_log /var/log/nginx/myhost.com.log combined; > try_files $uri =404; > #proxy_redirect off; > } > > ``` > > > > root@vps190138:~# cat /var/log/nginx-error.log > > > and it overrides log settings set on the main config level. > > > [mailman.nginx.org archive](http://mailman.nginx.org/mailman/listinfo/nginx) – Steve H. BSc(Hons) MIITP > **[Linux & Admin Support Services | "Services provided by people who are proud...](https://www.greengecko.co.nz/)** > > Offers linux support services, systems and database administration, consultancy and the full development lifecycle for open source - based solutions Linkedin: [Steve Holdoway - Robotron New Zealand Limited | LinkedIn](http://www.linkedin.com/in/steveholdoway) Skype: sholdowa ---
**Author:** ![Giulio\_L](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/g/0ea827/32.png) [@Giulio\_L](https://www.ruby-forum.com/u/Giulio_L)\ **Post date:** [March 5, 2016, 12:20am UTC](https://www.ruby-forum.com/t/tcp-connection/244203/10 "2016-03-05T00:20:28Z")
On Friday 04 March 2016 23:25:27 Giulio L. wrote: > ``` > access_log /var/log/nginx/myhost.com.access.log combined; > fastcgi_split_path_info ^(.+\.php)(/.+)$; > # NOTE: You should have "cgi.fix_pathinfo = 0;" in php.ini > > # With php5-fpm: > fastcgi_pass unix:/var/run/php5-fpm.sock; > fastcgi_index index.php; > include fastcgi_params; > fastcgi_param SCRIPT_FILENAME > > ``` $document\_root$fastcgi\_script\_name; > The file exists and it is fed as you can see below when I restart nginx: > > root@vps190138:~# cat /var/log/nginx-error.log > 2016/03/04 21:14:55 [notice] 13900#0: signal 3 (SIGQUIT) received, shutting > down > 2016/03/04 21:14:55 [notice] 13902#0: gracefully shutting down > 2016/03/04 21:14:55 [notice] 13902#0: exiting > 2016/03/04 21:14:55 [notice] 13902#0: exit > 2016/03/04 21:14:55 [notice] 13900#0: signal 15 (SIGTERM) received, exiting > 2016/03/04 21:14:55 [notice] 13902#0: signal 15 (SIGTERM) received, exiting > 2016/03/04 21:14:55 [notice] 13900#0: signal 17 (SIGCHLD) received > 2016/03/04 21:14:55 [notice] 13900#0: worker process 13902 exited with code > 0 > 2016/03/04 21:14:55 [notice] 13900#0: exit > 2016/03/04 21:14:55 [notice] 13949#0: using the “epoll” event method > 2016/03/04 21:14:55 [notice] 13949#0: nginx/1.8.1 > 2016/03/04 21:14:55 [notice] 13949#0: OS: Linux 3.13.0-66-generic > 2016/03/04 21:14:55 [notice] 13949#0: getrlimit(RLIMIT\_NOFILE): 1024:4096 > 2016/03/04 21:14:55 [notice] 13950#0: start worker processes > 2016/03/04 21:14:55 [notice] 13950#0: start worker process 13952 > root@vps190138:~# Do you have other configuration files in your /etc/nginx/conf.d/ and /etc/nginx/sites-enabled/ directories, which are included in your nginx.conf? wbr, Valentin V. Bartenev ---
**Author:** ![Giulio\_L](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/g/0ea827/32.png) [@Giulio\_L](https://www.ruby-forum.com/u/Giulio_L)\ **Post date:** [March 5, 2016, 7:02am UTC](https://www.ruby-forum.com/t/tcp-connection/244203/11 "2016-03-05T07:02:22Z")
Giulio, On 3/5/16 1:25 AM, Giulio L. wrote: > Here is my nginx.conf and site config. After calling tcp > connection and getting http 400, no log is generated. > > Im following Let me re-phrase: this is not your _full_ nginx config. There are probably other files in /etc/nginx or somewhere in your system that include the snippet below and/or other files. We know that because nginx just won’t start with the snippet below for a number of reasons. To diagnose the issue and provide advises we need full nginx configuration. If your nginx version (see nginx -V) is more or less recent, i.e 1.9.2+, please run “nginx -T” command to collect the configuration. Otherwise just find a folder with the nginx configs and check for another error\_log directives (e.g. grep -r error\_log /etc/nginx/). Hope this helps, Maxim – Maxim K.