For the record, the issue was accidentally assigning a mock User to Time.zone in a controller spec. Rails and RSpec would go in an endless loop of calling and recording a message.
On Thu, Nov 24, 2011 at 21:38, Marko Anastasov <ma...@renderedtext.com> wrote: > Hello, > > I'm having an issue running specs on a Rails project. Here's how > bundle exec rake spec --trace ends at a random point after some > seconds of hanging with a high CPU usage: > https://gist.github.com/1392208 > > I say "a random point" because I can run the problematic spec alone > without problems, and if I delete its content and run the entire suite > again the error will happen during another spec's example. Given that > doesn't change, it always repeats on the same example though. > > My setup is Ubuntu 10.04 on a Mac VirtualBox, bundler 1.0.18. Other > people in my team have the same setup (but not entirely the same ruby > patch level version etc), and do not have this issue. I'd very much > appreciate any hint how to solve this. _______________________________________________ rspec-users mailing list rspec-users@rubyforge.org http://rubyforge.org/mailman/listinfo/rspec-users
