The MPLS WG Archive

[Ina Minei <ina@juniper.net>]

	Alexa,

	The request to remove the Host FEC was proposed because it didn't
seem to be in use by any LDP implementation. Your input shows that this is
not the case.

	Luca, Arashmid, Markus, Rama,

	As proponents of this change, your input on this issue would be
appreciated.

	In particular, if the Host FEC is not deprecated, does it
still make sense to relax the rules of mapping packets to LSPs and treat
Host FECs and Address FECs equivalently when they are identical?

			Ina

On Thu, 7 Oct 2004, Alexa Morris wrote:

> This email is being sent on behalf of Rao Cherukuri, rcheruku@cisco.com.
>
> Dear George, Loa and Ina:
>
> A recent message on the MPLS WG mailing list
> (http://www.cell-relay.com/mhonarc/mpls/2004-Sep/msg00043.html) has
> suggested changes to RFC 3036 ("LDP Specification") and called for comments
> on the proposed changes.  One of the proposed changes is to deprecate the
> use of the Host Address FEC TLV.  Two Implementation Agreements published by
> the MPLS & Frame Relay Alliance (MFA), "MPLS Proxy Admission Control
> Definition" and "MPLS Proxy Admission Control Protocol", MFA.6.0.0 and
> MFA.7.0.0
> (http://www.mplsforum.org/tech/mpls-proxy-admission-control-definition-ia.pd
> f and
> http://www.mplsforum.org/tech/mpls-proxy-admission-control-protocol-ia.pdf)
> make use of the Host Address FEC TLV.  MPLS Proxy Admission Control provides
> a bandwidth-based reservation/admission facility for MPLS networks.  There
> are implementations of this protocol in progress.
>
> While the proposed changes make a Prefix FEC TLV with a 32-bit mask
> equivalent to a Host Address FEC TLV, adopting the Prefix FEC TLV would
> require changes to approved and published MFA documents, and changes to the
> implementations.  Also, since MPLS Proxy Admission Control only uses host
> addresses, using the Prefix FEC TLV would necessitate an additional check
> that the prefix was always 32 bits.  Therefore, the MFA kindly requests that
> the Host Address FEC TLV not be deprecated, and that it continues to be
> supported in future revisions of LDP.
>
> In RFC 3036, there is a semantic difference between a Host Address and a
> prefix with length 32.  The new version proposes to remove the semantic
> difference.  This is not a problem for the MPLS Proxy Admission Control; we
> are just requesting that the Host Address codepoint not be deprecated.
>
> Please advise us as soon as possible about the decision on this issue.
>
> Cordially,
> Rao Cherukuri
> MPLS & Frame Relay Alliance
> Technical Committee Chair
>
>
>
>

_______________________________________________
mpls mailing list
mpls@lists.ietf.org
https://www1.ietf.org/mailman/listinfo/mpls