[Libguestfs] [PATCH] spec: Relax NBD_OPT_LIST_META_CONTEXTS

[Wouter Verhelst]

Hi Eric,

On Fri, Sep 03, 2021 at 07:39:38AM -0500, Eric Blake wrote:
> Ping.
> 
> On Mon, Aug 16, 2021 at 01:40:59PM -0500, Eric Blake wrote:
> > Using OPT_SET_META_CONTEXTS is stateful (it is documented to wipe out
> > any previously-requested contexts, and we just tightened the spec to
> > clarify that starting TLS also wipes it out).  But
> > OPT_LIST_META_CONTEXTS is not stateful; and in fact, with a
> > SELECTIVETLS server, it can be handy to list the meta contexts
> > available on an unencrypted export, then enable encryption, and then
> > further list what contexts are available on encrypted exports (as the
> > server is permitted to let them differ).  Thus, while a wise client
> > will renegotiate structured replies after the starttls, there's no
> > reason to forbid a server from answering a client that uses
> > list_meta_contexts prior to encryption without also requesting
> > structured replies.
> 
> I originally wrote this patch prior to the point where we decided that
> OPT_STARTTLS should also wipe the effects of OPT_STRUCTURED_REPLY;
> given that change in the meantime, I'm tweaking that last sentence:
> 
> Although such a client must negotiate structured replies after
> starttls if it is going to actually connect to an export, this change
> permits the client to shorten the handshake by two commands if it is
> only being used to list available exports and their meta contexts.
> 
> > ---
> >  doc/proto.md | 4 ++--
> >  1 file changed, 2 insertions(+), 2 deletions(-)
> > 
> > diff --git a/doc/proto.md b/doc/proto.md
> > index 9dd59da..1586d7d 100644
> > --- a/doc/proto.md
> > +++ b/doc/proto.md
> > @@ -1325,9 +1325,9 @@ of the newstyle negotiation.
> >      Return a list of `NBD_REP_META_CONTEXT` replies, one per context,
> >      followed by an `NBD_REP_ACK` or an error.
> > 
> > -    This option MUST NOT be requested unless structured replies have
> > +    This option SHOULD NOT be requested unless structured replies have
> >      been negotiated first. If a client attempts to do so, a server
> > -    SHOULD send `NBD_REP_ERR_INVALID`.
> > +    MAY send `NBD_REP_ERR_INVALID`.

Yeah, makes sense. Setting a metadata context only makes sense once
structured replies have been negotiated, but we state that clearly
enough. Querying however doesn't set state, so that's fine.

-- 
     w at uter.{be,co.za}
wouter@{grep.be,fosdem.org,debian.org}