Hi Ole,
thx for your analysis of the problem space.
Please find my comments in line.
Ciao
Olaf
> -----Ursprüngliche Nachricht-----
> Von: Ole Troan [mailto:[email protected]] Im Auftrag
> von Ole Troan
> Gesendet: Montag, 13. September 2010 09:16
> An: Bonneß, Olaf
> Cc: [email protected]; [email protected]
> Betreff: Re: AW: New version available
>
> Olaf,
--------------
...
--------------
>
> so which solutions do we have:
>
> 1) only support DHCP on the BBF link-layer in the N:1 VLAN case.
> this will affect some legacy host implementations in the
> case where the customer has a bridged connection to the
> access network.
> how big a problem is this? considering that this is the
> route DOCSIS has gone with DOCSIS 3.0...
OBo: Agree that DHCPv6 could be used to provide address/prefix informations
also in N:1 VLAN scenario and you mentioned as well the affects to legacy host
implementations. I'm not speaking about DOCSIS networks.
> 2) RS mark.
> it appears consensus is that this will not work.
OBo: Hmm, I'm still not convinced that this will not work. May be I'm outside
the consensus here :).
> 3) tunneling and tagging traffic between AN and BNG.
OBo: Possible.
> 4) move L3 border to AN
OBo: Theoretically possible as well. But this means a heavy restructuring of
the whole access network infrastructure and the provisioning model etc.
> 5) change SLAAC
OBo: Extending SLAAC could provide a solution but it means also scratching the
fundaments of IPv6.
> with regards to the requirement that the solution should
> support existing host implementations that do not support DHCPv6.
>
> let us say we have 5 classes of host implementations:
>
> 1) IPv4 only and those which will never be upgraded with IPv6 support
> 2) partly broken IPv6 support and without DHCPv6
> 3) partly broken IPv6 support with DHCPv6
> 4) full IPv6 support without DHCPv6
> 5) full IPv6 support with DHCPv6
>
> by partly broken I mean lacking dual stack / IPv4/IPv6
> multihoming support. i.e happy eyeballs or having other
> serious short comings. I do not want to enable IPv6 on hosts
> implementations that e.g have 75 second timeout to switch
> from IPv6 to IPv4 (well known fruit vendor).
OBo: Hmm not sure if I understood what "partly broken" means. Have to think
about once more what you are trying to say here.
> we don't care about 1. we do _not_ want to deliver IPv6
> service to 2 + 3. so the problematic one is 4.
> does anyone know of any class 4 IPv6 implementation?
>
> cheers,
> Ole
>
>
--------------------------------------------------------------------
IETF IPv6 working group mailing list
[email protected]
Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------
