Re: [xri] RE: Delegation (was: trusted discovery workflow)

[Ben Laurie <benl@google.com>]

On Fri, Dec 5, 2008 at 7:10 PM, Breno de Medeiros <breno@google.com> wrote:
> And the endpoint would be signed because it is a legitimate endpoint
> for a different purpose?

The endpoint would not need to be signed because it is already in the
current authority (which was the initial one).

>
> Sounds to me like a defacement type of attack (mapping some services
> in the domain to others). Probably less interesting to do than
> straight defacement or DoS.

So, you are arguing that we should not defend against it because it is
less interesting than other attacks?

>
> On Fri, Dec 5, 2008 at 10:53 AM, Ben Laurie <benl@google.com> wrote:
>> On Fri, Dec 5, 2008 at 6:24 PM, Eran Hammer-Lahav <eran@hueniverse.com> wrote:
>>> This makes sense to me, but I think the challenge is how to codify this behavior in a way that is easy to understand and implement... It is way too easy to screw this up.
>>
>> It seems quite simple to me. At any point I have a current authority,
>> and until I hit an XRD signed by that authority, it doesn't change.
>> When I do, it becomes whatever the new XRD specifies. When I reach the
>> end, the current authority must be authoritative for whatever resource
>> I end up with.
>>
>> I do worry, though, that there could be some kind of attack where
>> (forged) unsigned delegations are used to point some service at the
>> wrong endpoint in the same domain, with the result that Bad Things
>> happen.
>>
>>>
>>> EHL
>>>
>>>> -----Original Message-----
>>>> From: Brian Eaton [mailto:beaton@google.com]
>>>> Sent: Friday, December 05, 2008 9:58 AM
>>>> To: Eran Hammer-Lahav
>>>> Cc: xri@lists.oasis-open.org
>>>> Subject: Re: [xri] RE: Delegation (was: trusted discovery workflow)
>>>>
>>>> On Fri, Dec 5, 2008 at 9:46 AM, Eran Hammer-Lahav <eran@hueniverse.com>
>>>> wrote:
>>>> > A domain delegating the management of their services to a third
>>>> party. Hosting and managing XRDs can and should become a product. Of
>>>> course, if you <ref> the entire XRD, you can as easily just point the
>>>> Link in that direction in the first place but it only works well with
>>>> Link header and element where you have a resource level control. If you
>>>> use a /site-meta map, it is impossible to point some XRD locations to
>>>> server A and some to server B (at least this is a use case I refuse to
>>>> support due to complexity). I much rather allow this to happen using a
>>>> simple <Ref> in the local XRD itself.
>>>>
>>>> This sounds like a reasonable use case, and I think it'll fit into the
>>>> trust workflow.
>>>>
>>>> - If the initial XRD is not signed or does not delegate to a new key
>>>> in the trust chain, the next XRD must be signed with a key
>>>> authoritative for the initial XRD.  More concretely: if I don't
>>>> specify that somebody else can sign an XRD for me, that next XRD needs
>>>> to be signed by me.
>>>>
>>>> - If the intial XRD is signed, it can point to a new key that will be
>>>> used to sign the next XRD.
>>>
>>> ---------------------------------------------------------------------
>>> To unsubscribe from this mail list, you must leave the OASIS TC that
>>> generates this mail.  Follow this link to all your TCs in OASIS at:
>>> https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php
>>>
>>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe from this mail list, you must leave the OASIS TC that
>> generates this mail.  Follow this link to all your TCs in OASIS at:
>> https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php
>>
>>
>
>
>
> --
> --Breno
>
> +1 (650) 214-1007 desk
> +1 (408) 212-0135 (Grand Central)
> MTV-41-3 : 383-A
> PST (GMT-8) / PDT(GMT-7)
>