-
-
Notifications
You must be signed in to change notification settings - Fork 925
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Intermittent NoMethodError in rspec #3959
Comments
Still having this; The stacktrace was:
NoMethodError: undefined method example_group_finished for NoMethodError:0x3291d9c2 |
StackOverflowError is considered fatal on JRuby, and we do not support recovering from it. FWIW, it's not safe to recover a StackOverflowError on any implementation, because it can happen at any time, in any code, including potentially VM code that then is unable to maintain a safe VM state. This is basically what's happening here. 9k may have improved this through a better codebase, better stack management, or more careful handling of VM state. In any case, the 1.7 line is no longer supported. If you are able to reproduce this on a recent JRuby 9k version and it does not involve stack overflow, we'll be happy to look into it. |
I filed rspec/rspec-core#2267 with the
rspec
project, but they were pretty confident it is a bug in JRuby so hence, rerouting here.This is with
rspec
3.4.1, as can be seen in the stack trace.I have no real idea on why this is happening, unfortunately. One of my specs is causing a
stack level too deep
error; could that be what's confusing it somehow?I did a repro case here: https://github.com/perlun/rspec-bug
It behaves differently with JRuby 1.7 and 9k. 9k never exposes this bug, but 1.7 does intermittently. Has something drastic changed regarding stack overflows on the 9k branch?
The text was updated successfully, but these errors were encountered: