k5_mutex_lock: Received error 22 (Invalid argument)

[Robbie Harwood]

Benjamin Kaduk <kaduk at mit.edu> writes:

> Hello,
>
> I do not believe this is a known issue.  The place that generates that
> diagnostic assert()s on failure, so hopefully you have one or more
> core dumps available.  The backtrace from the dump (ideally with line
> numbers in libkrb5support, but even without) would be a great place to
> start.

Right.

In particular, if the coredump suggests that the internal mutex state is
invalid, this points at memory corruption, and I would suggest
attempting to reproduce with valgrind.

(When I added the print statement, it was so that the compiler would
stop optimizing out this information.  At the time the underlying
corruption issue happened to be in 389ds, though this has of course been
fixed since.)

Thanks,
--Robbie
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 832 bytes
Desc: not available
Url : http://mailman.mit.edu/pipermail/krbdev/attachments/20201102/a8dbced0/attachment.bin