Mutex already locked error in pooling.rb (jruby/torquebox)

[Patrik Sundberg]

Hi,

I'm getting the following error at "random" times in an app I have deployed using jruby (1.6.7) on torquebox (2.0.1), using latest datamapper gems: https://gist.github.com/2507835

It's not something I can reproduce at will, seems to happen every few days under low load.

Anyone has an idea what's going on here? Bug in the pooling code?

Guessing jruby+TB may unearth some bugs under truly parallel threads not often encountered by most setups.

Thanks,

Patrik