Sign in to comment Contact GitHub API Training Shop Blog About © 2016 GitHub, Inc. Terms Privacy Security Status Help You can't perform that action at this time. Feb 12 17:37:16 dashboard-app-st-02 stderr.log: E, [2014-02-12T22:37:16.239518 #32444] ERROR -- : app error: deadlock; recursive locking (ThreadError) Feb 12 17:37:16 dashboard-app-st-02 stderr.log: E, [2014-02-12T22:37:16.239683 #32444] ERROR -- : /var/www/dashboard/shared/bundle/ruby/2.0.0/gems/rack-1.4.5/lib/rack/lock.rb:14:in `lock' Feb cannot be turned on in development mode. Source
You signed in with another tab or window. I've looked through a dozen or so popular middlewares and there is certainly no consistency in calling body.close. Maytag(tm) error puma/puma#360 Closed Relax Rack dependency rails/rails#9488 raggi added this to the Rack 1.5.3 milestone Jul 5, 2014 robotfelix commented Aug 8, 2014 +1 I've just started getting this issue websocket-rails member DanKnox commented Jan 21, 2013 That stack trace is not very helpful.
Skip to content Ignore Learn more Please note that GitHub no longer supports old versions of Firefox. However from the tracing through the code I have done it seems likely the problem exists with any method that passes responses through Rack::Lock. use Rack::ContentLength require ::File.expand_path('../config/environment', __FILE__) run FeaturesRails::Application It would be better if Faye closed the response. Why were Navajo code talkers used during WW2?
eprothro commented Jun 1, 2012 use ActionDispatch::Static use Bb::EndPoint use Rack::MobileDetect use Rack::Rewrite use Rack::Rewrite use Rack::Lock use Rack::Runtime use Rack::MethodOverride use Rails::Rack::Logger use ActionDispatch::ShowExceptions use ActionDispatch::RemoteIp use Rack::Sendfile use ActionDispatch::Reloader It seems to be working fine for me and at the very least it's better than what's currently there. Please try out the latest commits I just pushed and see if they fix your problem. bmxpert1 commented Jul 1, 2013 I have changed the code in my branch to match your sample, and I can confirm this works in Rails 4.
I don't know that Rails 4 is relevant, though @k1w1, I guess you're not using the thin-rails gem to make script/rails server use thin? So if a middleware below Rack::Deflater replaces the body in it's response, it will attempt to call close, and since those wrapper objects don't respond to it, the upstack body doesn't Personal Open source Business Explore Sign up Sign in Pricing Blog Support Search GitHub This repository Watch 118 Star 3,699 Fork 332 faye/faye Code Issues 9 Pull requests 3 Projects We don't know which exact request is causing it though.
if we do so we would need to release at some point :). No need for mutex w/ Thin. Any help much appreciated. As I am not sure what the issue could possibly be.
We recommend upgrading to the latest Safari, Google Chrome, or Firefox. bmxpert1 pushed a commit that referenced this issue Jun 26, 2013 Kevin Bouwkamp I can probably find a few minutes to try it out sometime next week (during the conference). This should solve your problem.
Faye sends the real response by calling env['async.callback'], possibly after a very long time, and Rack::Lock will not see this. http://crimsonskysoftware.com/unexpected-error/unexpected-error-while-processing-request-execution-expired.html gem 'rack', '1.4.1' With Rack 1.4.1, I also do not need to use Rack::ContentLength in config.ru. Someone reported issue was w/ airbrake gem, see at bottom. I've now removed Rack::Deflater and we haven't seen anything in the past 30 minutes.
You signed out in another tab or window. We hadn't identified a repeatable way to induce the error, so, we just yanked it. so that's not entirely perfect. http://crimsonskysoftware.com/unexpected-error/unexpected-error-while-processing-request-invalid-encoding.html Having a similar issue over here.
I've just updated @gregclermont's patch with another commit from master - https://gist.github.com/robotfelix/46758dc1c615e1908a1d This was referenced Aug 11, 2014 Closed Rack 1.5: app error: deadlock; recursive locking (ThreadError) with Rack::Deflater #721 Closed The error message I got was method_missing': undefined method threadsafe! –paulruescher Oct 20 '14 at 8:57 Yeah, this is no longer present in Rails 4. DanKnox closed this Jan 29, 2013 Sign up for free to join this conversation on GitHub.
This either needs to be middleware up the stack, that receives the response from Rack::Lock, or middleware down the stack, meaning it would have to call close on response after returning Though @danielgong I could reproduce on dev by clicking around our single page application a lot very quickly (perhaps firing multiple AJAX requests at the same time). bmxpert1 commented Jul 1, 2013 And yes that is why I haven't done a pull request, because I'm also not sure if that is the proper solution... Create your profile » People watching this ticket John Cant macournoyer Pages ↓↑ dragHome activereload/entp Our blog FAQ API Privacy Terms of service Copyright © 2007-2016 activereload/entp
jasonayre commented Jan 21, 2013 Ya Ive got it in the main config/application.rb file and double checked to make sure that none of the environments were overriding it. Mi cuentaBúsquedaMapsYouTubePlayNoticiasGmailDriveCalendarGoogle+TraductorFotosMásShoppingDocumentosLibrosBloggerContactosHangoutsAún más de GoogleIniciar sesiónCampos ocultosBuscar grupos o mensajes richard today Things that take my fancy today. Keeping the #unloadable call there for Rails 3 class reloading during development support is important so changing the implementation of the .inherited method in BaseController to the following should fix it: http://crimsonskysoftware.com/unexpected-error/unexpected-error-while-processing-request-end-of-file-reached.html What would have happened to the world if the sepoy mutiny of 1857 had suceeded?
The only thing is I don't know how to test this out; I'm not really sure what causes the problem to happen, it seems random and is much less frequent in I imagine that there might be something different about the Rackspace environment since the timing was too coincidental (and nothing else changed), but removing Rack::Deflater and gzipping via nginx fixes this Skip to content Ignore Learn more Please note that GitHub no longer supports old versions of Firefox. raggi added a commit that referenced this issue Jan 4, 2013 raggi