Ruby/JRuby and WEBrick serving requests one at a time

Question

Does anyone knows how to force WEBrick to process more than one request at a time? I'm using some Ajax on my page for long running database-related tasks and I can clearly see the requests are being processed in a pipeline.

Answer 1

If you use JRuby, check out the GlassFish gem (stripped-down GlassFish server in gem form), the Trinidad gem (same thing using Tomcat), or various other options like warbler (produces .war files you can run directly or deploy to any app server). JRuby is the easiest way for sure to deploy a highly-concurrent application on Ruby, and makes the C Ruby options look rather primitive in comparison.

Answer 2

webrick only processes one request at a time, which is usually fine for development. If you want things to run in parallel have a look at mongrel_cluster or the awesome unicorn or passenger of course.

Answer 3

You should definitely not be using WEBrick for long running requests. The best web server for the job would probably be Thin, as it's powered by EventMachine which makes it possible to write asynchronous code so that the server doesn't block.

Answer 4

If you are running rails 4 you can apply the following patch and webrick will start serving requests concurrently.

diff --git a/railties/lib/rails/commands/server.rb b/railties/lib/rails/commands/server.rb
index e3119ec..ef04aa8 100644
--- a/railties/lib/rails/commands/server.rb
+++ b/railties/lib/rails/commands/server.rb
@@ -93,14 +93,6 @@ module Rails
       middlewares << [Rails::Rack::Debugger]  if options[:debugger]
       middlewares << [::Rack::ContentLength]

-      # FIXME: add Rack::Lock in the case people are using webrick.
-      # This is to remain backwards compatible for those who are
-      # running webrick in production. We should consider removing this
-      # in development.
-      if server.name == 'Rack::Handler::WEBrick'
-        middlewares << [::Rack::Lock]
-      end
-
       Hash.new(middlewares)
     end