-
Notifications
You must be signed in to change notification settings - Fork 420
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
python segfault during test #650
Comments
I don't believe this is the same as #596: #596 related to the CFFI callback structure, and this backtrace is not coming out of a C callback. Instead it's coming out of What OpenSSL are you linking against please? |
|
Hrm, there's really nothing special about that OpenSSL: certainly not anything that should cause a segfault. Are you completely confident that your header files and library versions match? You may be using header files for a different version of OpenSSL than the |
Yes. That's the native version (NetBSD base), and no other version is installed. |
Hrm. It'd be good to know what test is executing. Can you run the tests in verbose mode ( |
|
Aha, ok, this may be callback related. That specific test validates that an exception being thrown in the private key passphrase callback is correctly handled. Segfaults would, IMO, not be considered "correctly handled". |
@reaperhulk, I'm going to page you in here: what's the state of affairs regarding #596? |
Is there an easy way to skip running this particular test to get a result for the other ones? |
Yeah, |
If I skip all tests that cause segfaults, the command line is:
the result of that is
Let me know if I should file separate bug reports for any of those. Thanks! |
Those should all be resolved outside of the segfault. #596 remains an issue because it appears libffi is buggy on NetBSD. Armin offered to help there, but he needs access to a system where he can replicate this so he can engage with the libffi maintainers if necessary. @0-wiz-0 if you have the ability to give him access to such a system then I'd recommend opening an issue on https://bitbucket.org/cffi/cffi and working with him. Otherwise we're stuck 😞 |
libffi is fixed, at least on NetBSD 8.99.1 with the latest libffi from pkgsrc, see libffi/libffi#294 |
cffi is fixed by @jsonn (see https://bitbucket.org/cffi/cffi/issues/321/cffi-191-segmentation-fault-during-self). I'll report other test failures separately. |
With 17.0.0, I see a segfault when running py.test. I don't know enough about py.test to identify the failing test, please advise.
Here's the output:
Here's the backtrace:
It might be the same error as #596, but it's not clear to me and the backtrace is much longer.
The text was updated successfully, but these errors were encountered: