[2015] in Kerberos_V5_Development
Re: Final issues before 1.0 code freeze
daemon@ATHENA.MIT.EDU (Theodore Y. Ts'o)
Mon Nov 25 13:19:16 1996
Date: Mon, 25 Nov 1996 13:18:17 -0500
From: "Theodore Y. Ts'o" <tytso@MIT.EDU>
To: Ezra Peisach <epeisach@MIT.EDU>
Cc: "Theodore Y. Ts'o" <tytso@MIT.EDU>, krbdev@MIT.EDU
In-Reply-To: Ezra Peisach's message of Mon, 25 Nov 1996 13:04:25 EST,
<9611251804.AA06928@kangaroo.mit.edu>
Date: Mon, 25 Nov 1996 13:04:25 EST
From: Ezra Peisach <epeisach@MIT.EDU>
I have confirmed the failure with OSF/1.
If KRB5_KTNAME is set, then the tests/dejagnu/gssapi.exp test fails.
I suspect that someone ran env_setup before make check and thats what
happened. I think adding the code from the gssftp test would solve
the problem.
OK, so this is probably an OS-independent problem, and as long as you
haven't set KRB5_KTNAME environment variable before running "make
check", you won't lose. Correct?
Adding an unsetenv of KRB5_KTNAME seems like the safest fix for now. In
the long term using something like the gssftp test is worth looking at,
although I think we need to take a long, hard look about the semantics
of KRB5_KTNAME, the kludge used by kadmind to set the value in kdc.conf,
and try to come up with a cleaner way GSSAPI applications can choose
which keytab file to use.
- Ted
