Re: Gen-ART Review of draft-ietf-forces-mib-07

[Robert Haas <rha at zurich.ibm.com>]

Spencer, thanks for the comments. See
below for the details. I reissued a draft with the editorial changes:

http://www.ietf.org/internet-drafts/draft-ietf-forces-mib-08.txt

Regards,
-Robert

"Spencer Dawkins" <spencer at wonderhamster.org>
wrote on 09/01/2008 04:07:03 PM:
> 
> Gen-ART Review of draft-ietf-forces-mib-07
> 
> Spencer Dawkins 
> 
> to:
> 
> ietf
> 
> 09/01/2008 04:10 PM
> 
> Cc:
> 
> "Ross Callon", "General Area Review Team", rha,
"Patrick Droz", 
> "Jamal Hadi Salim"
> 
> I have been selected as the General Area Review Team (Gen-ART)
> reviewer for this draft (for background on Gen-ART, please see
> http://www.alvestrand.no/ietf/gen/art/gen-art-FAQ.html).
> 
> Please resolve these comments along with any other Last Call comments
> you may receive.
> 
> Document: draft-ietf-forces-mib-07
> Reviewer: Spencer Dawkins
> Review Date: 2008-09-01
> IETF LC End Date: 2008-09-08
> IESG Telechat date: (not known)
> 
> Summary: This document is very close to ready for publication as a
Proposed 
> Standard. I have two technical comments below, but both are minor
issues 
> that could resolved in AUTH48 if you think they have merit.
> 
> Comments:
> 
> OBTW, idnits 2.08.11 runs cleanly (one spurious warning about [RFCzzzz]).
> 
> 3.  Introduction
> 
>    More specifically, the information in the ForCES MIB
module relative
>    to associations that are up includes:
> 
> Spencer (clarity): I'd suggest s/associations/associations between
Control 
> Elements and Forwarding Elements/, unless that's wrong (and if it
is, I'd 
> still suggest "between X and Y", whatever X and Y are).
Later uses of 
> "association" would be fine, of course, this is just providing
initial 
> context.
> 
> Spencer (clarity): I'd suggest s/up/in the UP state/, or at least
> all-caps-ing UP so it looks like a state.
> 

I made the suggested changes.

> 
> 4.  ForCES MIB Overview
> 
>    The MIB module contains the latest ForCES protocol version
supported
>    by the CE (forcesLatestProtocolVersionSupported).  Note
that the CE
>    must also allow interaction with FEs supporting earlier
versions.
> 
> Spencer (clarity): I'd suggest s/CE/Control Element (CE)/ (and same
for FE, 
> ID, etc.) The reader can figure stuff like this out, but spelling
it out for 
> the reader is just too easy.
> 

I made the suggested changes.

>    o  Number of other ForCES messages sent from the
CE
>       (forcesAssociationOtherMsgSent) and received
by the CE
>       (forcesAssociationOtherMsgReceived) since the
association entered
>       the UP state.  Only messages other than
Heartbeat, Association
>       Setup, Association Setup Response, and Association
Teardown are
>       counted.
> 
> Spencer (technical): I think I know what you're saying here, but you're
not 
> counting "other" messages (because you exclude some of the
"other" messages. 
> The point is that you didn't get into the table with Association 
> Setup/Association Setup Response, and you leave the table immediately
after 
> Association Teardown, so you don't have to count these messages, isn't
it? 
> :-(

I agree, but I'd rather keep this explicit. As for
"OtherMsg" vs "OperationalMsg": I'd rather keep it
as is, given that we define what are these "other" messages.

> 
> If it's not too late to change this to "OperationalMsg"
or something 
> similar, I'd suggest that, for clarity. If it is too late to change
this ...
> 
>    Finally, the MIB module defines the following notifications:
> 
>    o  Whenever an association enters the UP state,
a notification
>       (forcesAssociationEntryUp) is issued containing
the version of the
>       ForCES protocol running.  Note that as CE
ID and FE ID are
>       indexes, they appear in the OID of the ForCES-protocol
running-
> 
> Spencer (technical): this is intended as a question, because I don't
know 
> MIB practices, but it looks to me like CE ID and FE ID are concatenated
into 
> ONE index, so they aren't "indexes" - right? I'm looking
at the INDEX for 
> "ForcesAssociationEntry".
> 
> The rest of the document is pretty clear about this, so I'd suggest
"CE ID 
> and FE ID are concatenated to form the table index", or something
similar, 
> unless I don't understand (it happens).

Your interpretation is correct, I changed the text
as you suggested.

> 
>       version object.  Optionally, a notification
>       (forcesAssociationEntryUpStats) can instead be
issued with all
>       associated information for this association,
except
>       forcesAssociationTimeDown. 
> 
_______________________________________________
Ietf mailing list
Ietf at ietf.org
https://www.ietf.org/mailman/listinfo/ietf