Stewart,

I think it would be good to have the remote-LFA draft discuss and recommend
what addresses to use, what requirements there are on routers to make it
work (for IP or LDP), and what protocol gaps are identified.   What is
necessary for the PQ node to know to handle terminating an IP tunnel at
line rate?  What is needed to consistently pick the tLDP address, etc.?

Alia


On Thu, Dec 13, 2012 at 6:45 AM, Stewart Bryant <[email protected]> wrote:

> On 12/12/2012 16:34, Peter Psenak wrote:
>
>> Hannes,
>>
>> On 12.12.2012 17:05, Hannes Gredler wrote:
>>
>>  in favor of explicit advertisement, i'd rather do 3 rules here :
>>>
>>> 1. PQ node OSPF router-id, if it is advertised as /32 prefix by the PQ
>>> node itself
>>> 2. PQ node TE adress, if it is advertised as /32 prefix by the PQ node
>>> itself
>>>
>>> 3.  Highest /32 address advertised by PQ node in it's Router LSA
>>>
>>
>> ether (1) or (3) is mandatory for the t-LDP session creation. (2) is
>> optional and not sufficient for t-LDP session, so we can not have it before
>> (3).
>>
>> It looks to me we are trying to solve the configuration problem which
>> should not be addressed here.
>>
>
> I am inclined to agree, and note that the draft is tunnel independent.
>
> We may want a deployment guidelines document, or some protocol specific
> text in another document, but this draft can stand alone without this text.
>
> Stewart
>
> ______________________________**_________________
> rtgwg mailing list
> [email protected]
> https://www.ietf.org/mailman/**listinfo/rtgwg<https://www.ietf.org/mailman/listinfo/rtgwg>
>
_______________________________________________
rtgwg mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/rtgwg

Reply via email to