I had clarified such RAs in private discussions with various folks
inside and outside Cisco.

Sorry, if I have missed any email that said RA with no PIOs is viewed as
"incorrect". It is a totally correct RA configuration that 2461bis or
even RFC 2461 makes reference to. Now an RA that includes a PIO
advertising a prefix P with neither the L nor A bits, in absence of any
other sources of on-link information, is functionally no different that
an RA with no PIO  - both lead hosts receiving such RAs to send their
non-link-local traffic to default router. Soon as a host sends traffic
to default router, that is off-link operation. The other kind of RA that
Ralph mentioned is the second way one can configure an IPv6 router to
signal off-link behavior to a host. In an IPv6 network that is using a
router, and router is issuing RA to hosts, these are the only two ways
one can configure off-link mode in RA to signal to hosts. Hence for an
aggregation routed IPv6 network, these are two possible RA
configurations that the aggregation router can be configured for.

Here is justification from 2461bis, for the RA that has L bit clear and
advertising a prefix, that such an RA is signaling off-link. Note that
2461bis , in section 6.3.4 says, if L bit is clear, the RA is not
signaling off-link in the following text:

   [Note, however, that a Prefix Information option
   with the on-link flag set to zero conveys no information concerning
   on-link determination and MUST NOT be interpreted to mean that
   addresses covered by the prefix are off-link.  The only way to cancel
   a previous on-link indication is to advertise that prefix with the
   L-bit set and the Lifetime set to zero.  The default behavior (see
   Section 5.2) when sending a packet to an address for which no
   information is known about the on-link status of the address is to
   forward the packet to a default router;]

However, then the text above continues to say, "The default behavior
(see Section 5.2) when sending a packet to an address for which no
information is known about the on-link status of the address is to
forward the packet to a default router." Soon as 2461bis said "forward
the packet to default router", off-link operation is implied. That is
why the two RAs are not functionally different.

Hemant


-----Original Message-----
From: Ralph Droms (rdroms) 
Sent: Saturday, August 18, 2007 8:26 PM
To: IETF IPv6 Mailing List; [EMAIL PROTECTED]
Subject: Re: [dhcwg] Re: prefix length determination for DHCPv6

James - in the abstract, in my opinion the inference is that a prefix
cannot be described as on-link unless it is advertised in a PIO.

Question for the list: is there a functional difference between an RA
that includes a PIO advertising a prefix P with neither the L nor A bits
set (no on-link, don't do SLAAC) and an RA that does not advertise
prefix P at all?  I don't think there is, and I'm curious about why an
RA with no PIOs is viewed as "incorrect" while an RA with PIOs with L
and A not set is OK.

- Ralph

On Aug 17, 2007, at Aug 17, 2007,4:09 PM, james woodyatt wrote:


>
> On a related note, I've heard that some operators intend to deploy 
> DHCP service using RA with M=1 and no PIO.  I don't understand how
> they imagine the "on-link flag" to be propagated in that scenario.   
> The "on-link flag" seems to be clearly in the domain of router 
> function, not dynamic node configuration.  Is that to be described in 
> the forthcoming Internet draft?
>
>


--------------------------------------------------------------------
IETF IPv6 working group mailing list
[email protected]
Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------

--------------------------------------------------------------------
IETF IPv6 working group mailing list
[email protected]
Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------

Reply via email to