-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA512

David Holland wrote:
[---]
> One useful thing to find out would be whether, after things stop
> working, the process credentials stored in the kernel are still
> correct or if they've been garbaged. When things break, does running
> "id" print right or wrong information? Also, does ls -l (or
> stat(2)/fstat(2) if necessary(*)) return the right owner/group and
> permissions for the affected files?

   Ok, now I have verified two details I wasn't entirely sure about
previously.

   1) If I have a broken "pkgsrc" session (ls doesn't work), and I don't
close it, but rather use Ctrl+Alt+Fx to switch to another session, log
in/out with my "jan" user, and then switch back to the still active
"pkgsrc" session, then "ls" will start working. So it appears to be a
very global problem.

   2) Logging _in_ with my "jan" user doesn't make it work. It's the
logging *out* which makes the change to the better.

- --
Kind regards,
Jan Danielsson

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (NetBSD)

iD8DBQFG8IeIuPlHKFfKXTYRCplgAKCTysVxMhnDMeVucwlk5qO5tKjC2ACfeyGm
oT5vfP0NoNDew2mm0wPFUtc=
=rBRr
-----END PGP SIGNATURE-----