[dhcwg] *DRAFT* minutes from WG meeting in Seoul (2nd try)

[Ralph Droms <[e-mail]>]

Please review and respond with additions, corrections and deletions by
Friday, 3/26.

- Ralph

Minutes

Review of agenda.  No changes.

DHCP Option for Proxy Server Configuration
   <draft-ietf-dhc-proxyserver-opt-00>
   Technical discussion; ready for WG last call?

   Vijay gave a short review of his draft: this option sends a list of
   proxy servers and ports to a client.  ~8 WG members had read the
   draft.  The chair asked about the difference between this option and
   existing DHCP options for carrying lists of servers; for example,
   WWW and NNTP.  Vijay justified proxy server option because it
   provides port number as well as proxy server address.  The WG
   expressed consensus for WGLC (which will be confirmed on the WG
   mailing list).  Ted Lemon will publish editorial comments during
   WGLC.


The Extended Remote Boot Option for DHCPv4
   <draft-ietf-dhc-opt-extrboot-00>
   Technical discussion; ready for WG last call?

   Vijay gave a short review of his draft: this option gives a list of
   TFTP servers along with a list of files for each server.  ~7 WG
   members had read the draft.  There was discussion that this draft
   specifies multiple copies of the same option, with different
   semantics from those specified in RFCxxxx.  Also, the list of
   servers needs to differentiate between IP addresses or FQDNs.  There
   was a suggestion that the servers just be specified as IP addresses,
   and a suggestion that the file name be specified as a length
   followed by name string.  Another question: what are the semantics
   of the list of servers in case of an error?  Does the client skip to
   the next server, retry the file, or what?  Vijay will revise the
   draft to eliminate requirement for sub-options and address other
   comments.  There will be no WGLC at this time.


Vendor-Identifying Vendor Options for DHCPv4
   <draft-ietf-dhc-vendor-01>
   Technical discussion; ready for WG last call?

   The author has revised the draft based on discussion at last WG
   meeting.  ~8 WG members have read the draft.  There was one question
   about the vendor identifier, which was eventually recalled to be the
   enterprise number.  There was WG consensus for WGLC (which will be
   confirmed on the mailing list).


Node-Specific Client Identifiers for DHCPv4
   <draft-ietf-dhc-3315id-for-v4-01>
   Technical discussion; ready for WG last call?

   The author has revised the draft to simplify the generation of the
   DHCPv4 client identifier to match the DHCPv6 mechanism exactly.
   There was WG consensus for WGLC (which will be confirmed on the
   mailing list).


Rapid Commit Option for DHCPv4
   <draft-ietf-dhc-rapid-commit-opt-00>
   Technical discussion; ready for WG last call?

   The author gave a short presentation about this mechanism, which
   allows two message exchange in DHCPv4 (modeled on two message
   exchange in DHCPv6).  No comments from WG.  WG consensus for WGLC
   (which will be confirmed on the mailing list).  Ted Lemon to publish
   editorial comments during WGLC.


DHCPv6 Support for Remote Boot
   <draft-ietf-dhc-dhcpv6-opt-rboot-00>
   Technical discussion; ready for WG last call?

   The author gave a short presentation about this draft.  This option
   may be impacted by dual-stack problem resolution.  This option will
   not be affected by the multiple copies issue in the DHCPv4 option.
   Author to revise for editorial issues and for consistency (where
   possible) with similar DHCPv4 option.  No WGLC at this time.


Configured Tunnel End Point Option for DHCPv6
   <draft-ietf-dhc-dhcpv6-ctep-opt-01>
   Technical discussion; ready for WG last call?

   The author gave a short introduction to this option.  The WG
   expressed consensus that this option is intended for configuring
   routers, which is out-of-scope.  The Internet Draft will be dropped
   as WG work item (which will be confirmed on the mailing list).


Reclassifying DHCPv4 Options
   <draft-ietf-dhc-reclassify-options-00>
   Technical discussion; ready for WG last call?

   There was no comment from WG and the WG expressed consensus for
   WGLC (which will be configrmed on the mailing list).


DHCPv4 Support for Conf. IPv6-in-IPv4 Tunnels
   <draft-daniel-dhc-ipv6in4-tunnels-00>
   Accept as WG work item?

   The author gave a short presentation about the option.  ~8 WG
   members had read this draft.  There was some discussion of the draft
   on the mailing list prior to the WG meeting.  The author will
   publish a revised draft based on those comments. The draft will be
   reviewed by the v6ops WG; the chair will contact the v6ops WG chairs
   to determine how to proceed.  The WG also needs to confirm it fits
   in dhc WG charter.  The draft was not accepted as WG work item at
   this time.


Micro-block IP Addr. Alloc. With DHCP Proxy Server
   <draft-shen-dhc-block-alloc-01>
   Accept as WG work item?

   The author gave a short presentation about the option.  It fits in
   dhc WG charter under "address assignment".  The chair will arrange
   for a discussion to reconcile this option with DHCPv6 prefix
   delegation and earlier ODAP work.  The draft was not accepted as WG
   work item at this time.


Vendor-Specific Suboption for the DHCP RAIO
   <draft-stapp-dhc-vendor-suboption-00>

   Ted Lemon expressed concern that this option will lead to
   unstandardized RAIO sub-options.  The WG expressed consensus to
   accept as WG work item at this time (which will be confirmed on the
   mailing list).


Renumbering Requirements for Stateless DHCPv6
   <draft-chown-dhc-stateless-dhcpv6-renumbering-00>

   This draft defines requirements for updating configuration
   information obtained through the Information-Request message
   ("stateless DHCPv6").  There was a comment that the security sectino
   could be clearer.  The WG charter needs to be updated before the WG
   takes on this work.  The WG expressed consensus to accept this draft
   as a WG work item (pending charter update; will be confirmed on
   miling list).


Lifetime Option for DHCPv6
   <draft-venaas-dhc-lifetime-01>

   The author gave a short presentation about the draft.  This option
   is a solution for the renumbering problem. The chair asked if other
   conditions under which a host performs an Information-Request under
   other conditions like getting a new prefix in an RA.  The WG charter
   needs to be updated before the WG takes on this work.  The WG
   expressed consensus to accept this draft as a WG work item (pending
   charter update; will be confirmed on miling list).


Multicast Reconf. Protocol for Stateless DHCPv6
   <draft-vijay-dhc-dhcpv6-mcast-reconf-00>
   Accept as WG work item?

   The author gave a short presentation about the draft.  This option
   is another solution for the renumbering problem, which is especially
   applicable to unplanned changes.  The chair confirmed that there are
   two components: one that describes how hosts respond to the
   receonfigure message and one that describes how to trigger the
   router to send the reconfigure messages.  Margaret Wassserman will
   take some concerns to the dhcwg mailing list.  The WG compared this
   draft with the previous (lifetime option) draft; the WG consensus
   was that the two options are sufficiently different that they attack
   different problems.  The WG charter needs to be updated before the
   WG takes on this work.  The WG is very interested in this work, but
   ID was not accepted as WG work item.

(restart at 2:05hrs in)

IPv4 and IPv6 Dual-Stack Issues for DHCPv6
   <draft-chown-dhc-dual-stack-00>

   Stig Venaas gave an introduction to the problem based on
   draft-chown-dhc-dual-stack-00; this document also proposes some
   solutions.  kre remarked that title is all wrong; the draft really
   discusses problem of accepting configuration from multiple
   interfaces.  An alternative viewpoint is that information from two
   separate interfaces is different from having to make two answers to
   get all the information required.  It may be possible to solve the
   dual-stack problem without solving the more general problem of
   getting information from multiple sources.  The WG expressed
   consensus to accept this document as a WG work item (pending charter
   update; will confirm on mailing list).


Update of dhc WG charter

   "Par. n" refers to the nth paragraph in the current dhc WG charter.

   Par. 1 - OK
   Par. 2 - OK
   Par. 3 - work in progress (security); Droms to ask design team
            about progress
   Par. 4 - work in progress; design team (Hibbs, chair) reviewing RFC
            2131

   The WG agreed that the list of drafts will be dropped from the
     charter; inactive IDs to be explicitly dropped from WG work items

   New charter items:
   * Stateless DHCpv6 renumbering
     (<draft-chown-dhc-stateless-dhcpv6-renumbering-00>)
   * Dual stack issues <draft-chown-dhc-dual-stack-00>
   * DHCP proxy architecture - Narten to discuss with WG chair before
     adding to charter

   The WG chair will write up text for each of the three items and the
   WG will discuss the text on the mailing list

Update on IPR issue with two drafts

   This agenda item was overtaken events; WG to review three new RFCs:
   RFC 3667, RFC 3668, RFC 3669: and discuss IPR issues on dhcwg
   mailing list.