Re: [Webpush] Application server authentication new years edition

[Costin Manolache <[e-mail]>]

It is important to define what "voluntary" means :-). My understanding is
that it involves
a choice made by an entity, either the app server and push service.

The only 'voluntary' choice an app server has is to not use
a push service and UAs that requires authentication/authorization.

Separate comments on the 3 elements - authentication, authorization and
contact info.

For authentication:  2.1(certificate) would be my preference, and is a well
known and established
mechanism, followed by 2.6, 2.3.
-2.4 doesn't cover authentication,
-2.5 seems too complicated (2.5 maps to Oauth1, 2.3 to Oauth2 - we know the
outcome)
-2.2 is what we use now, but doesn't work well with 3. - subscription
association - if
public keys are used.

For authorization - or "subscription association" - big +1 :-)
I like using the public key of the sender - in which case authentication
must use 2.1, 2.6 or 2.3.
Using an ID provided by the push service will be complicated if multiple
push services are used.
I don't see any other option - whatever the subscriber uses needs to be
verifiable by the push service
when send happens.

For contact - each of the authentication schemes in the draft provide a way
to include contact info, and
the choice for senders is to include it or not, and the choice for push
services is to throttle/reject
 or not. The tricky part is if any additional verification will be done by
push services.
I guess some providers (in particular free services) may be ok with a more
relaxed verification
or even allow some rate-limited sending without contact info, I personally
don't mind it - if we
can't contact a sender in case of problems we just block it.

Costin

On Tue, Jan 5, 2016 at 7:35 PM, Martin Thomson <[e-mail]>
 wrote:

> With Peter's help, I've just submitted a new version of a draft that
> attempts to enumerate the options for application server
> authentication (issue #44).
>
>   https://tools.ietf.org/html/draft-thomson-webpush-vapid-01
>
> This assumes the following requirements:
>
> 1. A push service would like to be able to correlate requests from a
> particular application server over time.
>
> 2. A push service would like to be able to contact the
> operator/developer of an application server.
>
> This takes the position that any correlation or contact information is
> provided voluntarily by an application server and that we don't
> require strong authentication for application servers.
>
> Both Ben and Costin have suggested that we could (SHOULD?) construct a
> scheme to strongly authenticate servers.  Ben observed that most web
> push consumers will have HTTPS endpoints with valid server
> certificates and we could use that to construct a system that strongly
> authenticates senders.  That might be possible, but it's likely to be
> a complicated system that introduces a whole new set of technical,
> operational and privacy problems.
>
> The voluntary authentication incrementally improves the situation and
> allows us to build stronger authentication systems on top in future.
> I think that we should look at doing that, but I would be opposed to
> anything that made a more heavyweight authentication system mandatory
> to use.
>
> So I come to the question:  what levels of authentication do we want
> to support for application servers?
>
>   A. anonymous/none - push without being authenticated
>
>   B. voluntary authentication - as proposed in the draft
>
>   C. strong - proposal TBD
>


On Tue, Jan 5, 2016 at 7:35 PM, Martin Thomson <[e-mail]>
wrote:

> With Peter's help, I've just submitted a new version of a draft that
> attempts to enumerate the options for application server
> authentication (issue #44).
>
>   https://tools.ietf.org/html/draft-thomson-webpush-vapid-01
>
> This assumes the following requirements:
>
> 1. A push service would like to be able to correlate requests from a
> particular application server over time.
>
> 2. A push service would like to be able to contact the
> operator/developer of an application server.
>
> This takes the position that any correlation or contact information is
> provided voluntarily by an application server and that we don't
> require strong authentication for application servers.
>
> Both Ben and Costin have suggested that we could (SHOULD?) construct a
> scheme to strongly authenticate servers.  Ben observed that most web
> push consumers will have HTTPS endpoints with valid server
> certificates and we could use that to construct a system that strongly
> authenticates senders.  That might be possible, but it's likely to be
> a complicated system that introduces a whole new set of technical,
> operational and privacy problems.
>
> The voluntary authentication incrementally improves the situation and
> allows us to build stronger authentication systems on top in future.
> I think that we should look at doing that, but I would be opposed to
> anything that made a more heavyweight authentication system mandatory
> to use.
>
> So I come to the question:  what levels of authentication do we want
> to support for application servers?
>
>   A. anonymous/none - push without being authenticated
>
>   B. voluntary authentication - as proposed in the draft
>
>   C. strong - proposal TBD
>
> _______________________________________________
> Webpush mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/webpush
>