Re: [dhcwg] I-D Action: draft-ietf-dhc-dhcpv6-stateful-issues-06.txt

[神明達哉 <[e-mail]>]

At Mon, 30 Jun 2014 16:55:59 +0000,
"Bernie Volz (volz)" <[e-mail]> wrote:

> There is one outstanding issue that we'd like to get feedback on from folks (as well as of course the other changes made). We can discuss on the mailing list but will also discuss at the Toronto IETF DHC WG Session.
>
> This is the issue of which status codes Renew and Rebind should use when a server is unable to provide "leases" for a binding. Presently the text says to use NoBinding and part of that reason was that this is consistent with what is currently written in RFC 3315. But it may be better to use NoAddrsAvail/NoPrefixAvail instead of NoBinding. Marcin and I debated this some and here is some data regarding this issue:

Many points are discussed after these paragraphs, and I cannot easily
say which is the best with confidence.  But I figure out that it's at
least incompatible with existing client implementations if the server
returns something other than NoBinding.  According to RFC 3315, the
client will subsequently send a Request for the IA to (re-)establish
the binding.  If the server doesn't return NoBinding, and if the
client doesn't regard it as the need for some recovery action (sending
Request or Solicit), then the binding for the IA will eventually
expire (from a quick look at the code, this seems to be what would
happen with the WIDE DHCPv6 client implementation:
http://sourceforge.net/projects/wide-dhcpv6/).  It doesn't seem to be
a good idea to break backward compatibility this way unless there's a
strong reason for that.

BTW, as I read the entire draft it occurred to me that this draft
should probably describe what a client should do upon receiving a
Reply (in response to various types of messages) more specifically,
and with specific proposed text to revise Section 18.1.8 of RFC 3315.
Discussions on the client behavior on receiving a Reply will also be
related to the question on NoBinding vs others.

--
JINMEI, Tatuya