The MPLS WG Archive

Cell Relay Retreat>MPLS WG Archive>month:2001-Jun> msg00142



[Date Prev][Date Next][Thread Prev][Thread Next]  
  [Date Index][Thread Index][Author Index][Subject Index]

LDP's optional Transport Address TLV in the Hello Message

  • From: Hans Sjöstrand <hans@ipunplugged.com>
  • Date: Mon, 11 Jun 2001 12:06:52 +0200
  • Cc: <jcucchia@crescentnetworks.com>
  • Importance: Normal

Nobody raised the requirement to configure this specifically for LDP, so LDP
adheres to the general policies regarding wheither to use loopback or
local addreses.

For plattform label space I guess this is not an issue, and for interface
label space default is normally to use the interface addess.

I've seen the Transport Address TLV more as a mean to ensure that your
default-address-selection setting doesn't screw up your peer Transport
address
bindings. Not as a configurable option. This I belive is also the opinion
among the communiuty at large since the mib has been through some last
calls.
I'm reluctant to add new features to the mib at this late stage in the rfc
publishing process.

Regards
/// Hasse

> -----Original Message-----
> From: owner-mpls@UU.NET [mailto:owner-mpls@UU.NET]On Behalf Of Vach
> Kompella
> Sent: den 8 juni 2001 20:13
> To: mpls@UU.NET
> Subject: LDP's optional Transport Address TLV in the Hello Message
>
>
> Suppose I want this Transport Address TLV sent out in the Hello
> message.  If
> the default behavior is to use the loopback address, then which MIB table
> holds this info that there is an override Transport address?
>
> It is not in the Entity table.  How do I know, as I initiate
> Discovery that
> I should use this TLV?
>
> It's a per interface thing, so I guess there is no convenient
> place to hang
> it.  One possibility would be a field in the Entity that says
> mplsHelloTransportAddressFlag or something that is either enabled or
> disabled, so that either the interface or loopback address is used when
> Hellos start flying.
>
> I apologize if this has already been hashed out, but I couldn't find
> anything on it in the archives.
>
> -Vach