Re: [nfsv4] New Version Notification for draft-haynes-nfsv4-layoutwcc-00.txt

[Thomas Haynes <[e-mail]>]

On Feb 20, 2023, at 12:15 AM, David Noveck <[e-mail]> wrote:



On Sun, Feb 19, 2023, 6:53 PM [e-mail]<mailto:[e-mail]> <[e-mail]<mailto:[e-mail]>> wrote:
Hi Tom,
I like the idea. It looks like that the draft is limited only to NFSv3 based flexfiles layouts.

That was my impression as well.


Neither nfsv41_files nor flexfiles layouts allow GETATTR requests to be sent to data servers.

In retrospect, that was probably a design mistake in pNFS, but it's probably too late to address that now.


If we were to allow it, I would limit it to be after a COMMIT, READ, or a WRITE. I would not allow a compound like:

SEQUENCE
PUTFH
GETATTR

And simply such that we can mimic WCC, although it would now be SCC,



Thus only in responses to READ or WRITE over v3 clients will get the wcc information.

In nfsv4_files, the client could get them from the mds, with the understanding that the implementation would propagate this information from ds to the mds using the control protocol.

However the draft position itself to be more generic.

It does but that is probably the result of text written when this was part of delstid.  Most of delstid concerned general v4 extensions and the text might not have been appropriately adjusted to reflect the split.



No, it was a pure mental block on my end.



Do I miss something?

There is something missing but I feel it is the authors' responsibility to make this issue clear.


Yes. Lets see if the new draft clears it up.



Tom has requested we adopt this draft as a working group document.  Let's do that ASAP.  To make further progress, we need to address the scope issue you pointed out and the relationship of this document to the flexfiles layout document.

Best regards,
- Tigran
-------- Original message --------
From: Thomas Haynes
Date: Thu, Feb 16, 2023, 00:02
To: NFSv4
Subject: [nfsv4] Fwd: New Version Notification for draft-haynes-nfsv4-layoutwcc-00.txt


Hi,

As part of the review of delstid, it was suggested that I pull this part out of that document and create a new draft.

This then is that result.

I ask the working group to adopt this draft was a WG one. This is work that had previously been accepted as part of the delstid one.

Thanks,
Tom

Begin forwarded message:

From: [e-mail]<mailto:[e-mail]>
Subject: New Version Notification for draft-haynes-nfsv4-layoutwcc-00.txt
Date: February 15, 2023 at 2:58:00 PM PST
To: "Thomas Haynes" <[e-mail]<mailto:[e-mail]>>, "Trond Myklebust" <[e-mail]<mailto:[e-mail]>>


A new version of I-D, draft-haynes-nfsv4-layoutwcc-00.txt
has been successfully submitted by Thomas Haynes and posted to the
IETF repository.

Name: draft-haynes-nfsv4-layoutwcc
Revision: 00
Title: Add LAYOUT_WCC to NFSv4.2
Document date: 2023-02-15
Group: Individual Submission
Pages: 9
URL:            https://www.ietf.org/archive/id/draft-haynes-nfsv4-layoutwcc-00.txt
Status:         https://datatracker.ietf.org/doc/draft-haynes-nfsv4-layoutwcc/
Html:           https://www.ietf.org/archive/id/draft-haynes-nfsv4-layoutwcc-00.html
Htmlized:       https://datatracker.ietf.org/doc/html/draft-haynes-nfsv4-layoutwcc


Abstract:
  The Parallel Network File System (pNFS) allows for a file's metadata
  (MDS) and data (DS) to be on different servers.  It does not provide
  a mechanism for the data server to update the metadata server of
  changes to the data part of the file.  The client has knowledge of
  such updates, but lacks the ability to update the metadata server.
  This document presents several refinements to both RFC8881 and
  RFC8434 to allow the client to update the metadata server to changes
  on the data server.




The IETF Secretariat





_______________________________________________
nfsv4 mailing list
[e-mail]<mailto:[e-mail]>
https://www.ietf.org/mailman/listinfo/nfsv4
_______________________________________________
nfsv4 mailing list
[e-mail]<mailto:[e-mail]>
https://www.ietf.org/mailman/listinfo/nfsv4