[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <87eh272k5l.fsf@windlord.stanford.edu>
Date: Wed, 12 Mar 2014 14:15:34 -0700
From: Russ Allbery <eagle@...ie.org>
To: oss-security@...ts.openwall.com
Subject: Re: Cookie Reuse
Thomas Williams <thomas@...illiams.me.uk> writes:
> Would you consider the following scenario a security flaw:
> - User can login to an authenticated system
> - User copies the cookies that are set after they are authenticated
> - User logs out and the cookies are therefore deleted from their system
> - User can paste the previously copied cookies allowing them to re-auth
> without re-entering credentials
> Personally, I would feel that given the user has logged out, their
> expectation is that not only should the cookies be removed from their
> end, but essentially those cookies should no longer be accepted by the
> server at a later stage to re-authenticate.
Many bearer token systems have this security property, which is also
shared with (e.g.) Kerberos tickets. I think it's a security property to
be aware of, but I don't think it's a vulnerability unless the software
advertises otherwise.
Invalidating issued tokens prior to their expiration time requires storing
central session state or some revocation list equivalent, which may be an
undesireable or burdensome requirement for the security system as a whole.
Bearer tokens with expirations but without central session state
(Kerberos, for example) have some useful properties that cannot be easily
replicated by state-tracking systems, at the cost of making logout prior
to token expiration require destruction of all session keys.
--
Russ Allbery (eagle@...ie.org) <http://www.eyrie.org/~eagle/>
Powered by blists - more mailing lists
Please check out the
Open Source Software Security Wiki, which is counterpart to this
mailing list.
Confused about mailing lists and their use?
Read about mailing lists on Wikipedia
and check out these
guidelines on proper formatting of your messages.
