Re: [nfsv4] New Version Notification for draft-dnoveck-nfsv4-rpcrdma-xcharext-02.txt

[Chuck Lever <[e-mail]>]

Remarks on rpcrdma-xcharext-02.


- Credit accounting

There's no discussion of RPC-over-RDMA credit accounting in this
document. There needs to be some discussion of credit consumption.
In particular:

Requesters will have to have posted extra receive buffers (over
and above credits for forward channel replies and backchannel
requests) to deal with XCHAR messages. Likewise, responders will
need to post similar extra receives for this purpose.

Perhaps both peers should reserve one credit, and the specification
should insist that these operations are always single-threaded on a
connection.

Alternately, when xcharext is merged into rpcrdma-version-two,
there might be a generic discussion of non-RPC-payload-bearing
messages that could cover this issue.


- Overloading the term "initial"

Section 3 introduces an "initial set of transport characteristics"
and Section 4.1 defines "ROPT_INITXCH: Specify Initial
Characteristics". I think the use of "initial" means different
things in these two cases?

Instead:

Section 3 could propose "initially-defined" characteristics.

Section 4.1 could define ROPT_STARTXCH (like STARTTLS). I'm not
attached to that name, but it probably shouldn't be ROPT_INITXCH
if Section 3 defines "initial characteristics".

ROPT_CONNXCH

ROPT_FIRSTXCH

ROPT_EARLYXCH


- Section 4.4

The text in this section has been clarified to address previous
reviewer comments; thanks! There are a number of syntax and
grammatical errors that still need to be addressed (most often,
a few words are repeated in some sentences).

The "argument structure" of ROPT_UPDXCH is:

struct optinfo_updxch {
    xcharval        optupdxch_now;
    bool            optupdxch_pendclr;
};

I prefer optupdxch_new instead of optupdxch_now; "_now" suggests this
field records a time and/or date stamp.

It would help me to understand why a receiver needs to distinguish
between these types of notification.

For instance, if pendclr is false, this could be either a rejection
of a pending change request, or it could be an unsolicited change
notification. How does the receiver make use of the difference?

Instead of a boolean, an enumeration of update event types would be 
a little friendlier, and could be expressed in the same amount of
space (since an XDR boolean consumes 4 octets on the wire). Based
on the discussion in Section 4.4, we have:

enum optupdxch_event_type {
   OPTUPD_UNSOL  = 1,
   OPTUPD_MORE   = 2,
   OPTUPD_DONE   = 3,
   OPTUPD_REJECT = 4,
};

But since the rdma_xid field is not used to tie change requests to
these change update notifications, I'm not sure why the receiver
needs to know that a pending request has been completed. I think
REJECT might be more interesting than the difference between UNSOL,
DONE, and MORE.

There is a bit of a race here: a sender could send an unsolicited
update notification at the same time the receiver requests a change
of the same xchar. Could that result in a non-deterministic outcome?

Would it ever be reasonable to send two or more updates
simultaneously for the same XCHAR? (Requiring single-threading here
would prevent that from occurring).

What if the sender emits two optinfo_updxch messages: both with
pendclr set to false, but one with an intermediate value, and one
with the original value. The result on the receiver could depend on
the order in which these messages arrive. Possibly some text
regarding the ordering of these messages is needed.

What happens if the receiver of ROPT_REQXCH drops the request? Is
there a timeout after which ROPT_REQXCH may be sent again?

What happens if an ROPT_RESPXCH is dropped? If ROPT_REQXCH is sent
again, the reply is :

  ROPT_RESPXCH with the requested value marked done ?
  ROPT_RESPXCH with a rejection (no change was done) ?
  ROPT_UPDXCH with the requested value and pendclr set to true ?

I don't see language that disallows any of these responses. Which
one means "I already set this value" ? Sorry if I missed that.

Assuming that both sides support ROPT_UPDXCH, can an implementation
use ROPT_UPDXCH exclusively instead of ROPT_INITXCH?

Assuming that both sides support ROPT_UPDXCH, may a peer change an
XCHAR and not send an unsolicited ROPT_UPDXCH?


> On Aug 18, 2016, at 4:23 PM, David Noveck <[e-mail]> wrote:
> 
> This is updated and it add some vowels (and consonants too) the field and type names.  In particular "rq" --> "req".
> 
> I'm aware that some people find "XCHAR" confusing.  If someone has an idea for a replacement, please propose it on the list.  If the working group is OK with it, I'll produce a -03 incorporating it. 
> 
> 
> ---------- Forwarded message ----------
> From: <[e-mail]>
> Date: Thu, Aug 18, 2016 at 4:16 PM
> Subject: New Version Notification for draft-dnoveck-nfsv4-rpcrdma-xcharext-02.txt
> To: David Noveck <[e-mail]>
> 
> 
> 
> A new version of I-D, draft-dnoveck-nfsv4-rpcrdma-xcharext-02.txt
> has been successfully submitted by David Noveck and posted to the
> IETF repository.
> 
> Name:           draft-dnoveck-nfsv4-rpcrdma-xcharext
> Revision:       02
> Title:          RPC-over-RDMA Extension to Manage Transport Characteristics
> Document date:  2016-08-18
> Group:          Individual Submission
> Pages:          23
> URL:            https://www.ietf.org/internet-drafts/draft-dnoveck-nfsv4-rpcrdma-xcharext-02.txt
> Status:         https://datatracker.ietf.org/doc/draft-dnoveck-nfsv4-rpcrdma-xcharext/
> Htmlized:       https://tools.ietf.org/html/draft-dnoveck-nfsv4-rpcrdma-xcharext-02
> Diff:           https://www.ietf.org/rfcdiff?url2=draft-dnoveck-nfsv4-rpcrdma-xcharext-02
> 
> Abstract:
>    This document specifies an extension to RPC-over-RDMA Version Two.
>    The extension enables endpoints of an RPC-over-RDMA connection to
>    exchange information which can be used to optimize message transfer.
> 
> 
> 
> 
> Please note that it may take a couple of minutes from the time of submission
> until the htmlized version and diff are available at tools.ietf.org.
> 
> The IETF Secretariat
> 
> 
> _______________________________________________
> nfsv4 mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/nfsv4

--
Chuck Lever