RE: WG Last Call

Hi,

I agree that separate notifications are best.  The reason is that some
management systems like to differentiate between different events via the
top-level event definition rather than a varbind within it.  Also my gut
feeling is that "link fault" is a very important message, "dying gasp" could
be useful (but why isn't there a standardised "I'm dying" trap similar to
coldStart?) and "critical" is too vague to be usefully standardised.

Best regards,
	-- John

-----Original Message-----
From: [email protected] [mailto:[email protected]]On Behalf Of
Grant Schnebly
Sent: 16 August 2004 18:35
To: 'Matt Squire'
Cc: '[email protected]'
Subject: RE: [Hubmib] WG Last Call


Matt,

 > Do you think those notifications are sufficient ... or do you have a
different suggestion?

Please define a Notification for each distinct event.

-Grant



-----Original Message-----
From: Matt Squire [mailto:[email protected]]
Sent: Sunday, August 15, 2004 12:35 PM
To: John Messenger; [email protected]
Subject: RE: [Hubmib] WG Last Call



Hi John - 

Thanks for the comments.  Some replies below:

- Matt



> I have not reviewed the EPON or CU MIBs, but we have reviewed 
> and implemented the EFM MIB.  We agreed with one issue that 
> has already been raised:  as creation is automatic we couldn't 
> see the need for the rowStatus object.

Will be taken care of.

> Why is there no trap on receipt of OAMPDUs with the special 
> EFM alarm bits set in the flags field?  Assuming one was 
> reflecting the status of a remote system in a local system's 
> MIB, I can see some value sending such a trap.  In principle, 
> if there were suitable traps defined in other MIBs, I think 
> those should be used.  For the "link fault" event, I can 
> imagine using a trap in IF-MIB for example, but for a 
> "dying gasp" or "critical" event, there doesn't seem to be a 
> suitable trap elsewhere.  Arguably, the general-purpose "critical" 
> event is too generic to be useful, but there's real information in 
> "dying gasp" that hasn't been available from remote Ethernet 
> devices before.  We have chosen to include such events in traps 
> defined in an enterprise-specific MIB.

I tried to cover reflecting the flags in SNMP notifications using the
notifications dot3OamRmtErrEventFlags and dot3OamLclErrEventFlags.  These
notifications are supposed to be sent whenever the flags field changes
(which covers critical, dying gasp, link fault, both setting and clearing).
We could certainly pull those events out individually into their own
notifications, but I just threw it in as one for now.  

Do you think those notifications are sufficient, or that we should define
new notifications for each of the three flags, or do you have a different
suggestion?  

- Matt

-----Original Message-----
From: [email protected] [mailto:[email protected]]On Behalf Of
Romascanu, Dan (Dan)
Sent: 12 August 2004 14:55
To: [email protected]
Subject: FW: [Hubmib] WG Last Call


There is one more week until the WG Last Call completes. We received too few
comments to my taste. Please read the documents, and send your comments to
the WG list [email protected], if you want this work to be completed at the
appropriate quality level. Even comments like - 'I do not really have time
to review the MIB modules in detail, but the technical approach seems OK
from an EFM and Ethernet management perspective' are welcome. Of course, if
you think that something is broken, you should make a more explicit
argument.

Thanks and Regards,
Dan
-----Original Message-----
From: [email protected] [mailto:[email protected]]On Behalf Of
Romascanu, Dan (Dan)
Sent: 20 July, 2004 12:03 AM
To: [email protected]
Subject: [Hubmib] WG Last Call


My previous message contained an erroneous reference. Please disregard it,
and replace it with this one.
This is an Ethernet Interfaces and Hub MIB WG Last Call for the EFM MIB
Internet-Drafts. The documents can be found at: 
http://www.ietf.cnri.reston.va.us/internet-drafts/draft-ietf-hubmib-efm-mib-
01.txt
http://www.ietf.cnri.reston.va.us/internet-drafts/draft-ietf-hubmib-efm-epon
-mib-01.txt
http://www.ietf.cnri.reston.va.us/internet-drafts/draft-ietf-hubmib-efm-cu-m
ib-01.txt
All comments must be sent to the WG e-mail list [email protected], until
Wednesday August 18, 2004.
Although a 30 days Last Call period is being granted, it is strongly
suggested to the WG participants to submit their comments before the face to
face WG meeting, which will be hold on Monday, August 2nd - so that their
inputs can be discussed during the meeting in San Diego.


Regards,
Dan

_______________________________________________
Hubmib mailing list
[email protected]
https://www1.ietf.org/mailman/listinfo/hubmib

_______________________________________________
Hubmib mailing list
[email protected]
https://www1.ietf.org/mailman/listinfo/hubmib