[984] in Kerberos_V5_Development
Re: Krb5 & IP addresses
daemon@ATHENA.MIT.EDU (Sam Hartman)
Tue Jan 30 12:17:42 1996
To: "Richard Basch" <basch@lehman.com>
Cc: krbdev@MIT.EDU
From: hartmans@MIT.EDU (Sam Hartman)
Date: 30 Jan 1996 12:17:11 -0500
In-Reply-To: "Richard Basch"'s message of Tue, 30 Jan 1996 11:38:53 -0500
>>>>> "Richard" == Richard Basch <basch@lehman.com> writes:
Richard> I have a couple concerns with IP addresses being imbedded
Richard> within Kerberos 5.
Richard> I was talking with Ted about this, and it seemed that
Richard> there should be an easy way to simply say, "trust me from
Richard> any IP address". However, looking deeper into the code,
Richard> this is not the case.
I kind of got the impression from the rfc that the intent of encoding
no addresses in the ticket was to do this. I agree it doesn't work
all that well, considering that mk_safe and mk_priv apparently check
addresses.
Richard> I propose we add another address family type: AF_ANY,
Richard> ADDR_ANY, or something similar. kinit can then
Richard> optionally use that on the client to indicate that the
Richard> server need not check its originating address. This
Richard> would entail simply changing addr_comp.c and a couple
Richard> other functions.
This sounds like a poor solution to the problem, although I
can see reasons why it might be the best option available. (I'm
trying to think of a way to cleanly get the address for hostname
canonicalization purposes as I discussed yesterday, but to avoid using
it for comparison.) What are the obstacles to having a ticket with no
addresses encoded in it do what you want?
Richard> How does this sound?
Richard> Richard Basch Sr. Developer/Analyst URL:
Richard> http://web.mit.edu/basch/www/home.html Lehman Brothers,
Richard> Inc. Email: basch@lehman.com, basch@mit.edu 101 Hudson
Richard> St., 33rd Floor Fax: +1-201-524-5828 Jersey City, NJ
Richard> 07302-3988 Voice: +1-201-524-5049
