[nfsv4] Windows/NFSv4 ACL interoperability

[Spencer Shepler]

On Sun, J. Bruce Fields wrote:
> Several of us had a conversation about ACL interoperability at
> Connectathon the other week, and I just wanted to post some kind of
> summary.

[ snip big summary ]

I would like to thank Bruce for summarizing the discussion and bringing
it back to the WG alias.  Would also like to thank those that participated
in the discussion that followed.

Given that, there are elements of this thread of discussion that have
occurred on the NFSv4 WG alias before and to be honest, I have grown
weary.  I said as much in last week's WG meeting and suggested
a general course of action to avoid covering the issues again in the future.

In the NFSv4.1, we have included an expanded section describing the
NFSv4 ACLs.  This should be reviewed to ensure that we have a complete
description and agreement on the interpretations that have been drawn
from that.

Second, the NFSv4 WG will pursue 2 or 3 documents that will be
destined to be published as Information RFCs.  The first will be
a discussion of the mapping of the posix draft ACLs to/from the
NFSv4 ACLs.  This document has existed in the past and has been
dormant for some period.  It will be updated and republished and
reviewed by the working group.  The second document will be
a portion of the personal internet draft submitted by Sam Falkner
and Lisa Week.  That document was split such that the NFSv4 ACL
discussion was included in the NFSv4.1 document as mentioned above.
There was another section to the document that discussed the issues
of posix API behavior in the face of ACLs and specifically NFSv4 ACLs.
We have owners for those two documents.

Another potential set of documents would discuss the multi-protocol
filer environment (mixing NFSv4 and SMB ACLs) and then NFSv4 ACLs
in a SMB environment (similar to the posix API discussion).  We may
have an owner for this as well but will await confirmation.

Again, thanks for the input so far and please contribute to these
upcoming documents either directly or by review.

Spencer