Re: [nfsv4] I-D Action:draft-ietf-nfsv4-ipv6-00.txt

[Trond Myklebust <[e-mail]>]

On Mon, 2010-10-18 at 10:45 -0700, [e-mail] wrote:
> A New Internet-Draft is available from the on-line Internet-Drafts directories.
> This draft is a work item of the Network File System Version 4 Working Group of the IETF.
> 
> 
> 	Title           : NFS operation over IPv6
> 	Author(s)       : A. RN, et al.
> 	Filename        : draft-ietf-nfsv4-ipv6-00.txt
> 	Pages           : 8
> 	Date            : 2010-10-18
> 
> This Internet-Draft provides the description of problems faced by NFS
> and its various side band protocols, when implemented over IPv6 in
> various deployment scenarios.  Solutions to the various problems are
> also given in the draft and are sought for approval.
> 
> Foreword
> 
> This "forward" section is an unnumbered section that is not included
> in the table of contents.  It is primarily used for the IESG to make
> comments about the document.  It can also be used for comments about
> the status of the document and sometimes is used for the RFC2119
> requirements language statement.
> 
> A URL for this Internet-Draft is:
> http://www.ietf.org/internet-drafts/draft-ietf-nfsv4-ipv6-00.txt
> 
> Internet-Drafts are also available by anonymous FTP at:
> ftp://ftp.ietf.org/internet-drafts/
> 
> Below is the data which will enable a MIME compliant mail reader
> implementation to automatically retrieve the ASCII version of the
> Internet-Draft.

The RPCBIND requirement (MUST) is definitely unwelcome. The whole point
of giving NFSv4 a fixed port was in order to get rid of the reliance on
portmapping.

There is nothing in the document that justifies the prohibition on using
IPv4 embedded IPv6 addresses.

I don't understand the utility of the paragraph:
                While making a callback to an address received in a NLM
                LOCK call or a NFSv4 SETCLIENTID call, a server MUST
                specify the local interface via which the call needs to
                be made (or let the default zone be selected, if
                supported).
In NFSv4, if the server is calling the client back, then it is the one
initiating the TCP connection. There is no need for (nor is there even a
mechanism for) specifying on which interface the client must reply to
the RPC call.

Cheers
  Trond