Network Information Service (NIS) Configuration Options for Dynamic Host Configuration Protocol for IPv6 (DHCPv6)
draft-ietf-dhc-dhcpv6-opt-nisconfig-05

[Note]: 'This document was removed from the (extremely busy) 20-Nov agenda, when we found that it didn''t properly address Thomas'' AD review comments.  So, even if this ends-up in the "returning" section, this is really its first time through the full IESG.' added by Amy Vezza

[Ballot comment]
Here's the reference to NIS that I found.....

RFC 2307      Using LDAP as a Network Information Service    March 1998


  [XOPEN]
        ISO/IEC 9945-1:1990, Information Technology - Portable Operating
        Systems Interface (POSIX) - Part 1: Systems Application
        Programming Interface (API) [C Language]

[Ballot discuss]
what does a DHCPv6 server do when it wants to indicate that there's a NIS server available with an IPv4 address and another NIS server available with an IPv6 address?

The two reasons why this can be a silly question:

- NIS wouldn't work with such a config, so all the servers' addresses obviously have to be one address type
- You can always encode a v4 address as a v6 address using the 96-zeroes convention (or something else), so it's obvious how to do it

But I think the document should address the issue.

[Note]: 'This document was removed from the (extremely busy) 20-Nov agenda, when we found that it didn''t properly address Thomas'' AD review comments.  So, even if this ends-up in the "returning" section, this is really its first time through the full IESG.' added by Margaret Wasserman

[Note]: 'This document was removed from the (extremely busy) 20-Nov agenda, when we found that it didn''t properly address Thomas'' AD review comments.  So, even if this ends-up in the "returning" section, this is really its first time through the full IESG.' added by Margaret Wasserman

Determined that latest update does not resolve issues raised by Thomas Narten in May.  Author did respond indicating that Thomas' issues would be addressed.

From: Thomas Narten
To: vijayak@india.hp.com
cc: dhcwg@ietf.org
Date: Wed, 14 May 2003 15:48:17 -0400
Subject: [dhcwg] AD review of draft-ietf-dhc-dhcpv6-opt-nisconfig-02.txt

No big issues with this document, I went ahead and requested an IETF
LC. Here are my review comments.

NIS in abstract will likely need to be spelled out to satisfy rfc
editor..

>    This document uses terminology specific to IPv6 and DHCPv6 as defined
>    in section "Terminology" of the DHCP specification.

would be good to cite the RFC...

>    The format of the Network Information Service Servers option is as
>    as shown below:

s/as as/as/


>    The Network Information Service Servers option provides a list of
>    one or more IP addresses of NIS servers available to the client. The
>    NIS servers SHOULD be listed in the order of preference.

I think I understand the intent, but I think it would be more clear to
write this something like:

Clients MUST treat the list of servers as an ordered list.  Servers
(when they care) need to fill the fields in priority order (they are
not required to).

The above words don't really say the client MUST treat the option as
an ordered list.

Same wording issue appears in other places.

> 8. Appearance of these option

s/option/Options/ (also capitalize heading for consistency?)

> 9. Security Considerations
>
>    The NIS servers, NIS+ servers, NIS domain name and NIS+ domain name
>    options may be used by an intruder DHCP server to assign invalid NIS
>    parameters, resulting in clients unable to use NIS.

Is this really correct? it makes it sound like there is only a DOS
issue. Can't this option be used to send someoneto use a spoofed
server, in which case the server can provide bad responses that cause
errors or security compromises, or?

> 10. IANA Considerations
>
>    IANA is requested to assign an option code to these options from the
>    option-code space defined in section "DHCPv6 Options" of the DHCPv6
>    specification [1].

Would be good to list the options in this section together with TBDs
and pointers to the  sections. The way things are written now, IANA
has to scan the document in more detail to figure out what it needs to
do.

> 12. Informative References
>
>    [2]  Bradner, S., "Key words for use in RFCs to Indicate Requirement
>        Levels", BCP 14, RFC 2119, March 1997.

This references is generally considered normative.

[Ballot comment]
Informational reference [3] is currently:

  Sun Microsystems, "System and Network Administration", March 1990.

Surely there's a newer, better reference for NIS? How about the O'Reilly
book "Managing NFS and NIS, 2nd Edition"?