502 error nginx + ruby on rails application - mysql

Question

Application details :
Rails 3.1.0
Ruby 1.9.2
unicorn 4.2.0
resque 1.20.0
nginx/1.0.14
redis 2.4.8
I am using active_admin gem, for all URL's getting response 200,
but only one URL giving 502 error on production.
rake routes :
admin_links GET /admin/links(.:format) {:action=>"index", :controller=>"admin/links"}
And its working on local(development).
localhost log : response code 200
Started GET "/admin/links" for 127.0.0.1 at 2013-02-12 11:05:21 +0530
Processing by Admin::LinksController#index as */*
Parameters: {"link"=>{}}
Geokit is using the domain: localhost
AdminUser Load (0.2ms) SELECT `admin_users`.* FROM `admin_users` WHERE `admin_users`.`id` = 3 LIMIT 1
(0.1ms) SELECT 1 FROM `links` LIMIT 1 OFFSET 0
(0.1ms) SELECT COUNT(*) FROM `links`
(0.2ms) SELECT COUNT(count_column) FROM (SELECT 1 AS count_column FROM `links` LIMIT 10 OFFSET 0) subquery_for_count
CACHE (0.0ms) SELECT COUNT(count_column) FROM (SELECT 1 AS count_column FROM `links` LIMIT 10 OFFSET 0) subquery_for_count
Link Load (0.6ms) SELECT `links`.* FROM `links` ORDER BY `links`.`id` desc LIMIT 10 OFFSET 0
Link Load (6677.2ms) SELECT `links`.* FROM `links`
Rendered /usr/local/rvm/gems/ruby-1.9.2-head/gems/activeadmin-0.4.2/app/views/active_admin/resource/index.html.arb (14919.0ms)
Completed 200 OK in 15663ms (Views: 8835.0ms | ActiveRecord: 6682.8ms | Solr: 0.0ms)
production log : 502 response
Started GET "/admin/links" for 103.9.12.66 at 2013-02-12 05:25:37 +0000
Processing by Admin::LinksController#index as */*
Parameters: {"link"=>{}}
NGinx error log
2013/02/12 07:36:16 [error] 32401#0: *1948 upstream prematurely closed connection while reading response header from upstream
don't know what's happening, could some buddy help me out.

Answer 1

You have a timeout problem.
Tackling it
HTTP/1.1 502 Bad Gateway
Indicates, that nginx had a problem to talk to its configured upstream.
http://en.wikipedia.org/wiki/List_of_HTTP_status_codes#502
2013/02/12 07:36:16 [error] 32401#0: *1948 upstream prematurely closed connection while reading response header from upstream
Nginx error log tells you Nginx was actually able to connect to the configured upstream but the process closed the connection before the answer was (fully) received.
Your development environment:
Completed 200 OK in 15663ms
Apparently you need around 15 seconds to generate the response on your development machine.
In contrast to proxy_connect_timeout, this timeout will catch a server
that puts you in it's connection pool but does not respond to you with
anything beyond that. Be careful though not to set this too low, as
your proxy server might take a longer time to respond to requests on
purpose (e.g. when serving you a report page that takes some time to
compute). You are able though to have a different setting per
location, which enables you to have a higher proxy_read_timeout for
the report page's location.
http://wiki.nginx.org/HttpProxyModule#proxy_read_timeout
On the nginx side the proxy_read_timeout is at a default of 60 seconds, so that's safe
I have no idea how ruby (on rails) works, check the error log - the timeout happens in that part of your stack