Re: [rbridge] [Pppext] working group last call for PPP TRILL protocol control protocol [was Re: I-D Action:draft-ietf-pppext-trill-protocol-02.txt]

[Donald Eastlake <[e-mail]>]

Hi,

On Mon, Jan 24, 2011 at 5:44 PM, William Allen Simpson
<[e-mail]> wrote:
> On 1/24/11 3:41 PM, James Carlson wrote:
>>
>> William Allen Simpson wrote:
>>>
>>> I'm staring at the TRILL draft, and cannot find anything in the protocol
>>> to detect, eliminate, or resolve duplicate System ID numbers.  Obviously,
>>> I'm missing something.  Please advise.
>>
>> I don't believe that anything can or will do that.  It's assumed into
>> existence, as far as I understand.
>>
>> The System ID is used (among other things) for pseudonode generation and
>> TRILL nickname resolution.  If those aren't unique, then my
>> understanding is that the network will fly to bits.
>>
> There is some text about resolving nicknames that aren't unique.  Nicknames
> aren't deterministic on the System ID, though.  They're small, so they have
> a 2*8 average chance of collision.  (Section 3.7)

Nickname resolution is dynamic so collisions get resolved.

> I cannot find a description of pseudonode generation.  (Or even a
> definition for "pseudonode".)

That's because all of that is part of IS-IS. You need to start with
ISO 10589, the 2002 edition.

>> I'm not the IS-IS expert here.  If someone else (preferably on the
>> RBridge mailing list) has an opinion about running a TRILL IS-IS network
>> where some of the nodes using only point-to-point links have System IDs
>> that are possibly non-unique, then now is the time to speak up.
>>
>> But I certainly don't feel comfortable endorsing this sort of usage in
>> the PPP TRILL draft without review by at least the TRILL group, and
>> probably the IS-IS group as well.
>>
> OK, let's wait a bit.
>
> The questions are:
>
>  - How does TRILL handle System IDs that are not unique?

System IDs are used to uniquely identify IS-IS router in the link
state database. There is no TRILL specific stuff related to non-unique
System IDs. It's all IS-IS.

>  - Are System IDs ever used in the ethernet source or destination fields?

There is no requirement to ever to that. Since they are the same size
you can. As far as I know (which isn't that much), most manufacturers
of IS-IS switches either allocate an MAC address they control to use
as the System ID for a box or, a bit more commonly, use the MAC
address of one of the boxes interface, such as the numerically lowest
interface MAC address.

There is no requirement for the MAC interfaces of RBridges in a campus
to all have unique addresses as long as all the RBridges connected to
a particular link have unique MAC addresses over that link. But the
System IDs must be unique across the campus as per IS-IS subnet
independent functions requirement.

>> I see nothing in RFC 1661 section 6.4 requiring or even suggesting a
>> check across all of the node's interfaces.  Off hand, I don't know of
>> any that do that, and I do know of a few implementations where such a
>> test would be architecturally impractical (if not outright impossible).
>>
> The whole point is to avoid loops.  It is "a unique number".  You cannot
> avoid a loop with your *own* interfaces without checking against any
> existing interfaces to see whether your chosen number is unique....
>
> Certainly, I'd never expect connecting serial port 1 to serial port 2
> on the same machine would ever pass the magic number test.
>
> (Sometimes I wonder, how many more details must we write in specifications,
> when something this obvious is missed by some implementers?)
>
>
>> With Ethernet, the assumption is that MAC addresses among systems that
>> speak IS-IS are unique per the standards.  Whether any vendor is able to
>> achieve that and/or whether relying on MAC uniqueness is a smart thing
>> to do is possibly an interesting topic, but I think is certainly one for
>> a different mailing list.  It doesn't involve PPP.
>>
> I don't like assumptions.  That's a tautology.
>
> If the proven existing MAC uniqueness in the field is already *less* than
> known mathematical uniqueness of the PPP Magic Number, then there's no
> PPP-specific issue to be solved.  In either case of conflict, then manual
> configuration is required.

See above. MAC address uniqueness isn't the same as System ID
uniqueness although in practice they might be conflated sometimes.

Donald

_______________________________________________
rbridge mailing list
[e-mail]
http://mailman.postel.org/mailman/listinfo/rbridge