I used the mentioned document to help out with determining the tuple of
(epoch, cid, callNumber) for uniquely identifying the client's request.
It was extremely helpful in that regard.

I may have missed where it specifies that the response must come back to
the client in a packet that contains that exact same tuple.

I have looked through it (as mentioned below) and the reference documents.

I have still not seen any hints as to how an AFS request-reponse pair can
be identified.   Any help from the community would be greatly appreciated.

Looking back through the document, I see that:

Rx Connections
==============

        "epoch is a unique value chosen by Rx on startup and used by the
        peer to both to identify connections to this host"

        The information regarding connection ID (cid) and the 4 channels
        is present.

        "Call number identifies a particular call within a channel"

        "Sequence numbers always start with 1 at the beginning of each
        call."

        CLIENT-INITIATED is indicative of client initiated packet and lack
        of it indicating a server packet.

Packet Types
============

        DATA type packets have an Rx payload (operation)

Call Layer
==========

        A new call is established by the client simply sending a
        data packet to the server on an available channel

I have also reviewed all of the associated documents mentioned at the end
of the document, to no avail.

Any assistance from the community would be much appreciated.


On Fri, 2 Nov 2001, Nickolai Zeldovich wrote:

> > How do I know when a packet is a sent by a server in response to a
> > particular client's request?
>
> You might want to take a look at this writeup of the Rx spec:
>
>   http://web.mit.edu/kolya/afs/rx/rx-spec
>
> which should clarify most such questions, hopefully.  Let me
> know if there's something missing from the spec that should
> be there.
>
> -- kolya
> _______________________________________________
> OpenAFS-devel mailing list
> [EMAIL PROTECTED]
> https://lists.openafs.org/mailman/listinfo/openafs-devel
>


--Andrew J. Klosterman
[EMAIL PROTECTED]
http://www.ece.cmu.edu/~andrew5

_______________________________________________
OpenAFS-devel mailing list
[EMAIL PROTECTED]
https://lists.openafs.org/mailman/listinfo/openafs-devel

Reply via email to