phc-discussions - Re: [PHC] Transforming hash to different cost setting

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]

Date: Sat, 22 Mar 2014 15:47:52 +0100
From: Krisztián Pintér <pinterkr@...il.com>
To: discussions@...sword-hashing.net
Subject: Re: [PHC] Transforming hash to different cost setting

Thomas Pornin (at Saturday, March 22, 2014, 2:33:38 PM):

> there is a need to regularly (not often, but regularly)
> increase the hash cost,

it is not that tight. upgrading every year seems perfectly
appropriate.

>  - It requires the login process to be able to process two kinds of
> hashed passwords,

not if we are talking about parameter upgrade. you can't upgrade
between different primitives anyway. and being able two use different
parameters does not look very hard.

> implies that there must be some extra old/new flag stored along

or the parameters themselves. note that this is also required if you
want upgradable hash, unless you can upgrade all of them in one go
while logging in is disabled.

>  - It requires the login process to contain the re-hash logic. This
> is not hard, but that's still extra code and complexity

seems negligible to me.

>  - Rehashing upon login means that the login system must pay the
> cost for the old hash AND the new hash.

rehashing is spread out in a year or more. for ever rehash, there will
be dozens if not hundreds of regular logins. the relative cost should
be irrelevant.

> - The upgrade process is then spread over time; a given password
> will be upgraded only when the owner comes by next, which can take a
> few months or years.

which is perfectly fine. that also spreads out the load. if the use
does not come back for a long time, the password hash should be
destroyed.

> it has its own cost, namely
> that the user, if he evers come back, will need to go through the
> hoops of the forgotten password process.

which, again, is a negligible cost compared to the usual routine. we
are talking about users that didn't show up for at least a year.

> As a rule, users don't like these password things;

they also don't like their stuff be stolen, mails read, etc. it is all
about education. and if they still prefer weakly hashed passwords over
the hassle, a service provider can simply skip upgrading. after all,
we want to serve the people, not the cryptographers. if people don't
want safe passwords, fine. all we can do is to make it available.

but again, i can't imagine it would cause a significant user
dissatisfaction if they face some password reset procedure after
coming back to a website after a year of absence. i can't imagine
angry forum posts about it. personally, i always forget my password
after such long time, so i start with resetting anyway.

> "Offline cost upgrade" is a feature that some password hashing
> functions may provide.

the question is not whether we want to use it if present. the question
is, how much value it represents. my point is that not much. two
candidates needs to offer very much the same, for upgradeability to be
the deciding factor.