Re: [amqp] Review: Message Annotations For Response Routing Version 1.0 (WD02)

[Keith Wall <kwall@redhat.com>]

A couple of comments:

Section 2.5, Message Rewriting

- in sentence "One alternative mechanisms [if] for the intermediary to
rewrite the request message." the  if ought to be is.
- in sentence "Further the node at the rewritten address will need to
convert any references to the message-id of the [rewritten] message to
a reference to the message-id of the original message (e.g. in the
correlation-id property)" I think the word "rewritten" should be
"response".  Also "the node at the rewritten address" seems awkward.
Is this trying to indicate the "rewriting node"?

Section 2.3 uses "responseaddresssupported"  where as Section 3 uses a
hyphenated form "response-address-supported"

Front page, Working Draft number disagrees with the document's footer.








On Wed, Oct 9, 2019 at 11:05 AM Rob Godfrey <rgodfrey@redhat.com> wrote:
>
> I've re-reviewed this document [1] and think there are a few trivial items that need to be addressed before progression to CSD / CSPRD:
>
> Section 2 Response Annotations, first paragraph following "Let us consider an example:"
>   The last sentence prior to the bullet points states "G_A is listening on an address T.." - It should instead say something like "G_A has established a receiving link from target T in the remote network" ... the term "listening" has no defined meaning in AMQP.
>
> Section 2.2 Target Capabilities, paragraph following the table.
>   "If a target does not support response annotations, then a which carries the response [...]" - this should read "If a target does not support response annotations, then a message which carries the response [...]"
>
> Section 2.3 Delivery Annotations (Request Message)
>   In the definition for the annotation name "response-address-cookie-expiry".  First sentence "If present this delivery annotation [...]" should read "If present, this delivery annotation".  In the same paragraph the last sentence currently reads "After this point in time messages sent with the address-cookie annotation set to the value of the response-address-cookie should be expected to be rejected."  Do we all agree the a disposition of rejected is correct, and if so what error would we expect to be carried with this?
>
>
> Given the above comments I think we need a new working draft before progressing to CSD/CSPRD.  Does anyone else have any comments on this document?
>
> -- Rob
>
> [1] https://www.oasis-open.org/committees/document.php?document_id=65466&wg_abbrev=amqp
> --
> _____________________________________________________________________________
>
> Red Hat GmbH, www.de.redhat.com,
> Registered seat: Grasbrunn, Commercial register: Amtsgericht Muenchen, HRB 153243,
> Managing Directors: ,Charles Cachera, Michael Cunningham, Michael O'Neill, Eric Shander
>