[208] in Kerberos-V5-bugs
Re: another telnetd bug fix for krb5.
daemon@ATHENA.MIT.EDU (Theodore Ts'o)
Mon Oct 28 11:18:26 1991
Date: Mon, 28 Oct 91 11:17:56 -0500
From: tytso@Athena.MIT.EDU (Theodore Ts'o)
To: Rocky - Rakesh Patel <rapatel@theophilus.rutgers.edu>
Cc: krb5-bugs@Athena.MIT.EDU
In-Reply-To: Rocky - Rakesh Patel's message of Mon, 28 Oct 1991 00:46:49 -0500 (EST),
Reply-To: tytso@Athena.MIT.EDU
Hmm.... it seems like if there are bugs with the assumptions in the
authentication code, they should be fixed instead of kluging around the
code in start_login(). When is startslave() being called with
autoname set to NULL? I seen one call in telnetd.c which has the wrong
number of arguments which is set off by an #ifdef convex; that needs to
be fixed, but I don't think that was your problem.
- Ted
P.S. My apologies for the poor quality of the telnetd code; the problem
is that too many people have been modifying it and adding #ifdef's for
their particular configuration. The readibility of the code is fast
approaching nil. The fact that the authentication code is called out of
getterminaltype() is unforgiveable....
