Brian Haberman wrote:
IPv6 WG, One of the issues raised during IESG review of
draft-ietf-ipv6-over-ppp-v2 was the lack of clarity in the
implementation reports on the support of compression, which is
required in order to move the specification to DS. Calls went out
within the PPP and compression communities for additional reports
detailing the interoperability of this portion of the spec. Those
calls were unanswered.
Rather than re-cycle the entire spec at PS, Jari has asked that the
specification be split. The compression option will move to a new
draft targeted as a Proposed Standard. The remaining components will
be progressed to Draft Standard. The current draft editor, Srihari
Varada, has agreed to edit both documents.
Comments, questions, or concerns?
Thanks for asking. I have just discovered this document exists, and I'm
happy with it.
I have several comments, but they're not to slow the progress. If too
complex to address I can raise them after DS.
Thanks,
Alex
While the document does talk about stateless address autoconf I think it
should also talk about ND. I think it should mention whether - or not -
it's possible, feasible, has been done or probable, to run IPv6 ND over
a ppp link. If yes, then the SLLAO/TLLAO options (source link-layer
address option, and target) could be encoded from the Interface IDs
derived by IPv6CP. And if we talk link-local IPv6 multicast addresses
on the ppp links then also we can have a meaningful way to describe them.
As long as the Interface-Identifier is negotiated in the IPV6CP phase
of the PPP connection setup, it is redundant to perform duplicate
address detection (DAD) as a part of the IPv6 Stateless Address
Autoconfiguration protocol [3] on the IPv6 link-local address
generated by the PPP peer. It MAY also be redundant to perform DAD
on any global unicast addresses configured (using an
Interface-Identifier that is either negotiated during IPV6CP or
generated, for instance, as per [9]) for the interface as part of the
IPv6 Stateless Address Autoconfiguration protocol [3] provided that
the following two conditions are met:
1) The prefixes advertised, through the Router Advertisement
messages, by the access router terminating the PPP link are exclusive
to the PPP link.
2) The access router terminating the PPP link does not autoconfigure
any IPv6 global unicast addresses from the prefixes that it
advertises.
Not sure I understand the 2nd condition here. Is it that we don't want
the AR to _statelessly autoconfigure_ (a router never does anyways) or
that we don't want the AR to have an address in same prefix as the host?
How would absence of global address on AR help avoiding DAD?
I also think since ppp is used to connect to an Access Router that that
AR should be a default router, i.e. we can state it in terms of ND RA
lifetimes.
Still in the section stateless autoconfig, it would be worth mentioning
whether the initial RA is sent to all-nodes link-local address, or to
the link-local address that the PPP server delivered to the host.
Finally, there are ppp implementations that don't negotiate the IID with
the server (non compliant with this spec) but who'll still print a
link-local IPv6 address, fe80::something. That "something" is always
the same, on all hosts here or in Australia, and it's not 0. It allows
the IPv6 stack so still somehow work even if the PPP peer is not IPv6
capable and if this spec isn't used. If you think it's worth talking
about this non-compliant IPv6 ppp interface then I could dig the form of
that 'something'.
Thanks for the draft.
Alex
--------------------------------------------------------------------
IETF IPv6 working group mailing list
[email protected]
Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------
