phc-discussions - Re: [PHC] Password hashing by itself is not enough

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CAOLP8p6ZaMNRoq3EJZ1x5Zy-Fs3wR1mEiZj0yQ49RKKFuJ-Oag@mail.gmail.com>
Date: Fri, 13 Mar 2015 14:43:01 -0700
From: Bill Cox <waywardgeek@...il.com>
To: "discussions@...sword-hashing.net" <discussions@...sword-hashing.net>
Subject: Re: [PHC] Password hashing by itself is not enough

On Fri, Mar 13, 2015 at 7:26 AM, Thomas Pornin <pornin@...et.org> wrote:

> From all of the above, I recommend the following:
>
>  -- Publish a PHS "standard API" that does not include any provision for
>  an additional secret key.
>
>  -- Publish _another_ API, possibly even an _implementation_, that wraps
>  around the previous one, and provides key-based password hashing with
>  the "method 3": PHS is used to produce a 128-bit output, and that output
>  is encrypted symmetrically with AES (single-block encryption).
>
>  -- Write a note that states that an additional key is _possible_, and
>  if one wants it then it should be done properly (as demonstrated by the
>  implementation from the previous point), but cannot be recommended in
>  all generality because whether it is a good idea or not depends on the
>  context. Keys mean key management, an often underestimated thorny
>  issue.
>
> Arguably, doing anything with such key strengthening is outside of the
> scope of PHC and may/should be left to other people.
>
>
>         --Thomas Pornin
>

I 100% agree with this recommendation.  Sorry for this lame "+1" post, but
it's getting buried among other comments.  A second API, an implementation,
and some warnings about the difficulty and dangers of key management sound
about right to me.

Bill

Content of type "text/html" skipped