Minutes IETF106: lsvr

Meeting Minutes Link State Vector Routing (lsvr) WG Snapshot
Title Minutes IETF106: lsvr
State Active
Other versions plain text
Last updated 2019-11-20

Meeting Minutes


a. Welcome (10min)
Update on WG status provided by WG co-chair Gunter.

b. Update of Shortest Path Routing Extensions for BGP Protocol (15min)
   Presented by Keyur Patel
Keyur: based on feedback received, added capability to mark nodes not capable
of acting as a transit node. Alvaro: YANG model exists ? Keyur: working on it,
not ready yet.  Also working on implementation and protocol experience drafts
Alvaro: What is the protocol implementation draft ? Keyur: specific information
such as inter-op. Acee: wanted to keep it simple, not put typical IGP bells and
whistles in it.  Open to feedback on other functionality that is required.  
Protocol experience draft: it's a requirement in order to publish a new
protocol spec.

c. Update Layer 3 Discovery and Liveness (10 min)
   by Randy Bush
Randy: 4th version and has not changed much.  Primary goal is topology
discovery for L3.  Thinking of moving to WGLC ? Acee: Suggested that is should
go to WGLC to prevent it getting bigger. Randy: it has not changed *that* much
Tony P: Have you considered DCCP ? Randy: does not make sense since you don't
know the IP addresses yet.

d. Update Layer 3 Discovery and Liveness Signing (10 min)
   Presented by Randy Bush
Randy: requesting WG adoption

e. Update L3DL Upper Layer Protocol Configuration (10 min)
    Presented by Randy Bush
Randy: requesting WG adoption

f. Update LSVR IETF Organizationally Specific TLVs for IEEE Std 802.1AB (LLDP)
   Presented by Paul Congdon
Sue Hares: individual draft in idr on TLV usage for LLDP
Randy: how often will msgs be exchanged ?
Paul: will be covered in the next update (LLDPv2)

g. Update LSVR IEEE P802.1ABdh (==LLDPv2) (15min)
   (Informational IEEE Liaison work Update)
   Presented by Paul Congdon
Acee: this work is happening independently of lsvr ?
Paul: yes, also used for other work such as 802.1Qcj
Keyur: feels that both this and the L3DL work should proceed in parallel
Paul: not proposing this as an exclusive method
Alvaro: confirming that both Keyur and Randy said that 802.1ABdh and L3DL are
two ways of achieving the same thing.  Does 802.1ABdh satify requirements that
we are addressing with L3DL ? Keyur: these are two different transports and
semantics using the same TLVs.  Believes that the choice should be open. Randy:
they are *not* the same TLVs but semantics need to be the same, which is not
clear at this point. Alvaro: can we define the TLVs once and use them in both
protocols ? Paul: packing of TLVs may be different. Possible to define
identical TLVs .... Randy: LLDPv2 has a real-estate problem in terms of TLV
size... L3DL does not.

h. Wrap-up