[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [nfsv4] FedFS Meeting Minutes, 10/22/2009

James,
These topics have been discussed in the past. I'm trying to refresh my memory on where we left things. Two questions:
1) Did we ever work on a doc describing the root fileset details (database schema, etc)? 2) How do file servers exporting the root fileset export both the unified namespace root fileset, and other exports? Are we expecting a separate IP for a server to export "/" of the root fileset, so the physical path of other exports don't appear in the namespace? 

--Paul

On 2009-Oct-22, at 11:40, James Lentini wrote:



FedFS Meeting Minutes, 10/22/2009
---------------------------------

Attendees
---------

Craig Everhart (NetApp)
Sorin Faibish (EMC)
James Lentini (NetApp)
Robert Thurlow (Sun)

Minutes
-------

+ IETF Note Well Agreement

 This is a reminder that our discussions are governed by the
 IETF Note Well Agreement. See:

   http://www.ietf.org/NOTEWELL.html

 We will start each week's meeting with this announcement.

+ Draft Updates

 The IETF website's NSDB and Admin drafts are now several months old.
The plan is to update the drafts on the IETF website with the changes we have accumulated before the IETF draft update cutoff on Monday 10/26. We plan to make at least one further update to the drafts in mid- November after 
 the IETF'76 meeting.

+ NSDB Draft Update

 The working version of the NSDB draft is here

  http://jlentini.users.sourceforge.net/draft-ietf-nfsv4-federated-fs-protocol-04.txt

 and a diff against the -03 version is here:

  http://jlentini.users.sourceforge.net/draft-ietf-nfsv4-federated-fs-protocol-rfcdiff.html

 The update includes:

 * updated boilerplate for pre-RFC5378 contributions
 * Removed NFS-specific FSL fields from the overview and concepts
   section. With the number of NFS-specific fields growing, the
   overview was becoming drowned in details.
 * Changed "NSDB location" and "NSDB server" to "NSDB node" for
   consistency. The "NSDB node" term is what we define in the
   glossary, use occasionally in the NSDB draft, and use in the
   requirements document.
 * Clarified examples in Section 3 (Nico requested this on the mailing
   list)
 * Added the NSDB Container Entry concept to allow flexible LDAP
   configurations (Nico requested this on the mailing list)
 * Removed text about the conventional DN of the privileged LDAP user
   (cn=admin,o=fedfs). Nico recommended this on the mailing list.
 * Added CODE BEGINS/CODE ENDS markers to LDAP schema to clearly
   indicate the license on these definitions.
 * Defined a fedfsNfsPathname to be an XDR encoded field. There
   are concerns about viewing and editing this field to discuss.
 * Split fsl_info into separate attributes for flag bits, class,
   order, and rank fields. This allows searches on these individual
   attributes.
 * Listed the references to the FedFS admin protocol and FedFS
   requirements as informational. Neither are required to implement
   the NSDB protocol and the requirements draft, as an informational
   document, cannot be a normative reference.
 * Added tracking FSN references as an example use of annotations
* Stated that an FSL's validFor (time a client may cache a referral) and 
   TTL (time a server may cache a referral) may be different.
* LDAP UID space partitioned more logically with 1-99 for generic attributes, 
   100-199 for NFS attributes, 1000+ for object classes
 * NFS FSL format doesn't contain attribute for FSLI4GF_CUR_REQ or
   FSLI4GF_ABSENT. These will be set by the fileserver. Should the
   document say something about this?

 TODO: Use of DNS SRV for locating an NSDB

 James tested the new schema in OpenLDAP and OpenDS. As expected, both
 handled the new attributes correctly.
Sorin asked if the "NSDB node" term was clear. He said he would review the document and suggest changes if he felt clarifications were necessary. 

+ Admin Draft Update

 The working version of the Admin draft is here

  http://jlentini.users.sourceforge.net/draft-ietf-nfsv4-federated-fs-admin-03.txt

 and a diff against the -02 version is here:

  http://jlentini.users.sourceforge.net/draft-ietf-nfsv4-federated-fs-admin-rfcdiff.html

 * updated boilerplate for pre-RFC5378 contributions
 * updated pathname definition to match NFSv4 format
 * added NSDB Container Entry value to FSN

 TODO: Add recommended operations for setting NSDB Trust Anchors

 We have 2 options to provide the above functionality:

 - Add optional operations to the Admin protocol
 - Recommend the use of an existing (or soon to exist) protocol

 The pkix WG is chartered to work on this:

 http://www.ietf.org/dyn/wg/charter/pkix-charter.html

 and has produced the following:

 Trust Anchor Management Requirements
 http://www.ietf.org/id/draft-ietf-pkix-ta-mgmt-reqs-04.txt

 Trust Anchor Format
 http://www.ietf.org/id/draft-ietf-pkix-ta-format-04.txt

 Trust Anchor Management Protocol (TAMP)
 http://www.ietf.org/id/draft-ietf-pkix-tamp-03.txt

_______________________________________________
nfsv4 mailing list
nfsv4 at ietf.org
https://www.ietf.org/mailman/listinfo/nfsv4