Libguestfs - Aug 2021 - [PATCH] spec: Relax NBD_OPT_LIST_META_CONTEXTS

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.
---
 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`.

     Data:
     - 32 bits, length of export name.  
-- 
2.31.1

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`.
> 
>      Data:
>      - 32 bits, length of export name.  
> -- 
> 2.31.1
> 
-- 
Eric Blake, Principal Software Engineer
Red Hat, Inc.           +1-919-301-3266
Virtualization:  qemu.org | libvirt.org