Re: [HACKERS] SCRAM authentication, take three

On Fri, Apr 7, 2017 at 9:59 AM, Heikki Linnakangas <hlinnaka@iki.fi> wrote:

On 04/07/2017 10:38 AM, Magnus Hagander wrote:

So here's a wild idea. What if we just call it "sha256"? Does the user
actually care about it being scram, or is scram just an implementation
detail for them? That way when the next one shows up, it'll be sha512 or
whatever. It happens to use scram under the hood, but does the user have to
or does the user want to care about that?

(One could argue the same way that the user shouldn't have to or want to
care about the hashing algorithm -- but if that's the case then we should
only have one entry, it would be "scram", and the system would decide from
there. And I think this discussion already indicates we don't think this is
enough)

I think the "SCRAM" part is more important than "SHA-256", so -1 on that.

If that is the important part, then I agree :) I am not entirely sure that the scram part *is* more important though.

    I agree it is much more important. Needed, I'd say. "SHA-256" could refer to other mechanisms that just simply hash the value (maybe with a salt, or not) with that hash algorithm. SCRAM is a different beast, with much more functionality than that. So yes, it matters a lot :)

I think most users will be a lot more comfortable with "sha256" than "scram" though. But I guess that says using scram-sha-256 is the correct way.

    I don't like UPPERCASE, but the RFC links to the IANA registry where SCRAM methods are all uppercase and with dashes: SCRAM-SHA-256 and SCRAM-SHA-256-PLUS. I'd use those names, they are standardized.

 

The main against using just "scram" is that it's misleading, because we implement SCRAM-SHA-256, rather than SCRAM-SHA-1, which was the first SCRAM mechanism, commonly called just SCRAM. As long as that's the only SCRAM variant we have, that's not too bad, but it becomes more confusing if we ever implement SCRAM-SHA-512 or SCRAM-something-else in the future. That's the point Noah made, and it's a fair point, but the question is whether we consider that to be more important than having a short name for what we have now.

Yeah, I agree we should be prepared for the future. And having "scram" and "scram-sha-512" would definitely fall under confusing.

 

The channel binding aspect is actually more important to think about right
now, as that we will hopefully implement in the next release or two.

In [1], Michael wrote:

There is also the channel binding to think about... So we could have a
list of keywords perhaps associated with SASL? Imagine for example:
sasl    $algo,$channel_binding
Giving potentially:
sasl    scram_sha256
sasl    scram_sha256,channel
sasl    scram_sha512
sasl    scram_sha512,channel
In the case of the patch of this thread just the first entry would
make sense, once channel binding support is added a second
keyword/option could be added. And there are of course other methods
that could replace SCRAM..

It should also be possible to somehow specify "use channel binding, if the
client supports it".

Is that really a type of authentication? We already hvae the idea of
authentication method options, used for most other things except md5 which
doesn't have any. So it could be "sha256 channelbind=on", "sha256
channelbind=off" or "sha256 channelbind=negotiate" or something like that?

> Technically, the channel-binding variant is a separate SASL mechanism, i.e. it has a separate name, SCRAM-SHA-256-PLUS. I'm not sure if > users/admins think of it that way.

I bet they don't.

    Probably. But let's not underestimate channel binding: it is the "greatest" feature of SCRAM, and where security really shines. I'd encourage the use of channel binding and would definitely make it explicit.

    As for the options, there's no way to negotiate, the client picks. It could still be three-valued: on, off, only-channel-binding (or however you want to call them). That will only change what mechanisms the server will be advertising to clients.



    Álvaro

-- 

Álvaro Hernández Tortosa


-----------
<8K>data