krb5-23 intermittently fails on OSX.

[magnumripper]

Is this a Heimdal bug in OSX?

[kholia]

How do I reproduce this problem?

I can't say if I will be able to look at this problem soon enough.

[magnumripper]

I get it in unstable and in bleeding, with OMP and without it. It never fails when running --test --format=krb5-23, only when running just --test or -t=0. And sometimes it doesn't fail :-)

[magnumripper]

I do think it's a bug in the lib. We can just document it in doc/BUGS and ignore it.

[magnumripper]

http://www.openwall.com/lists/john-dev/2013/04/29/42 seems to confirm this is a lib bug.

[kholia]

Documented in 110ef7b. Closing.

[magnumripper]

It is not documented in that commit nor in any other commit (yet).

[kholia]

My bad.

After carefully looking at the BUGS file, I now see that it only mentions krb5-23 failing on big-endian CPUs. It doesn't say anything about krb5-23 failing on OS X.

[magnumripper]

For some reason I haven't seen this problem in a week or two now.

[magnumripper]

Doh, I saw it again now.

[magnumripper]

Documented in 0fc2c73