Re: [Idr] draft-ietf-idr-aigp-11.txt

[Thomas Mangin <[e-mail]>]

Hello Eric,

Following your changes, I have no more feedback (at least atm - never say never :p) on this draft.
Good luck with its progression.

Thomas

On 21 Nov 2013, at 15:44, Eric Rosen <[e-mail]> wrote:

> I have revised this draft per the WG Last Call comments:
> 
> - Error handling for a malformed AIGP attribute is "discard attribute".
> 
> - An AIGP attribute with the transitive bit set is regarded to be malformed,
>  and is to be discarded.
> 
> - Stated explicitly that the AIGP value used in bestpath computations is the
>  value of the first AIGP TLV in the attribute.
> 
> - The requirement to default "AIGP_SESSION" to "enabled" on IBGP sessions
>  has been demoted from "MUST" to "SHOULD".  This reflect the fact that
>  there is some controversy about it, but no real consensus to default it to
>  "disabled".
> 
> - Stated that the AIGP value is capped at its maximum (unsigned) value,
>  i.e., it should not wrap around due to addition.
> 
>  Also stated that if an AIGP attribute is received and its value is already
>  the maximum, the attribute should be treated as malformed.  (It's not
>  really useful if you can't add to it.)
> 
> I would like to thank those of you who commented during the last call, and
> would welcome comments on the changes.
> 
> 
> 
> 
> _______________________________________________
> Idr mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/idr
>