Re: [nfsv4] persistent sessions and compound atomicity

["J. Bruce Fields" <[e-mail]>]

On Wed, Dec 09, 2020 at 07:46:40AM -0500, David Noveck wrote:
> On Tue, Dec 8, 2020, 7:26 PM J. Bruce Fields <[e-mail]> wrote:
> 
> > I hadn't given much thought to this language in RFC 5661, and now that I
> > do, it surprises me.  If you have any pointers to discussion I forgot or
> > overlooked, I'd be interested:
> >
> 
> I don't know of any but. I have comments that might be helpful.
> 
> 
> >         https://tools.ietf.org/html/rfc5661#section-2.10.6.5
> >
> >         The execution of the sequence of operations (starting with
> >         SEQUENCE) and placement of its results in the persistent cache
> >         MUST be atomic.
> 
> 
> Although this is ambiguous, I don't think the intention here is to say the
> execution of the Compound needs to be a atomic.  The text is citing two
> things that must be atomic.  It would normally be reasonable to infer that
> each of them must be atomic.  However, the difficulties of doing that,
> given that most individual ops don't need to be atomic. means I think it is
> not intended meaning.  An alternative interpretation is that the transition
> between an existing Compound and a completed compound must be atomic.
> 
> If a client retries a sequence of operations
> >         that was previously executed on the server, the only acceptable
> >         outcomes are either the original cached reply or an indication
> >         that the client ID or session has been lost (indicating a
> >         catastrophic loss of the reply cache or a session that has been
> >         deleted because the client failed to use the session for an
> >         extended period of time).
> >
> 
> In my view, the atomic transition is between encountering an executing
> compound and getting the cached reply for a completed compound.

OK, got it, I think that makes sense.

--b.