Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 17 Oct 1996 22:25:14 +0400 (MSD)
From:      "=?KOI8-R?Q?=E1=CE=C4=D2=C5=CA_=FE=C5=D2=CE=CF=D7?=" (Andrey A. Chernov) <ache@nagual.ru>
To:        pst@shockwave.com (Paul Traina)
Cc:        CVS-committers@freefall.freebsd.org, cvs-all@freefall.freebsd.org, cvs-lib@freefall.freebsd.org
Subject:   Re: cvs commit: src/lib/libskey skey_getpass.c
Message-ID:  <199610171825.WAA02128@nagual.ru>
In-Reply-To: <199610171704.KAA17144@precipice.shockwave.com> from "Paul Traina" at "Oct 17, 96 10:04:20 am"

next in thread | previous in thread | raw e-mail | index | archive | help
> I'm not sure this is a good idea.  You want to keep things "stealthy"

I already think about it. It very is different from
'don't ask password for nonexisten user' case, because
you can even visually detect situation when any password
will be incorrect.

You'll see following picture if you can't match with _any_ password:

	login: user
	(s/key required)
	Password:

It is a bit different from normal case:

	login: user
	s/key 99 xx1234
	(s/key required)
	Password:

I.e. if "s/key ..." line is missing, it is pretty clear
that it is impossible to enter, so removing password asking
not makes it less "stealthy".

For FTP case the same thing:

	331 S/key password required for user.

Is a bit different from normal

	331 s/key 99 xx1234

response.

--------------------------------------------------------------------

BTW, as alternate solution another way is possible, I mean:

o Remove "(s/key required)" line if user can't specify any password
  and still ask for them.

o Change 'S/key password..." to usual "Password required..."
  for FTP responce in the same case and still ask for password.

This method makes even your s/key system to be undetected
for 1) wrong user names or 2) users not in skey database.
It is clearly good for 1) but not so good for 2), so I not decide yet.
			  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Making it for 1) but not for 2) is bad way because allows to determine
which user name is valid.

S/key already introduce some flaw in this direction, I mean
that you receive "s/key 99 xx1234" responce for users from skey database
and NOT receive it for users not in the database. So, it can be
possible to determine it from outside. We can
produce fake "s/key <rnd> xx<rnd>" response, but it pretty
well detected for repeatable access to the same name.

Any ideas?

-- 
Andrey A. Chernov
<ache@nagual.ru>
http://www.nagual.ru/~ache/



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199610171825.WAA02128>