RE: AD Review: draft06 draft-ietf-ipcdn-pktc-mtamib-06.txt ( Comm ent 13)

 
The authors will keep the resolution proposed below for Comment #13 unless
there are other suggestions from the WG.
 
Eugene.

________________________________

From: Wijnen, Bert (Bert) [mailto:[email protected]] 
Sent: Tuesday, September 13, 2005 1:07 AM
To: Eugene Nechamkin; Wijnen, Bert (Bert); Jean-Francois Mule; Ipcdn (E-mail)
Cc: Richard Woundy @ Comcast
Subject: RE: AD Review: draft06 draft-ietf-ipcdn-pktc-mtamib-06.txt ( Comm
ent 13)


I personally still prefer two objects (AdminSTatus and OperStatus), but I can
live with your solution and so it is uo to the WG to speak up if they see
an isseu.
 
Bert
 

	-----Original Message-----
	From: Eugene Nechamkin [mailto:[email protected]]
	Sent: Tuesday, September 13, 2005 00:05
	To: Wijnen, Bert (Bert); Jean-Francois Mule; Ipcdn (E-mail)
	Cc: Richard Woundy @ Comcast
	Subject: RE: AD Review: draft06 draft-ietf-ipcdn-pktc-mtamib-06.txt (
Comm ent 13)
	
	
	
	 
	> Maybe add some of that explanation to teh DESCRIPTION clause then.
	 
	PROPOSED: modify the description clause to the following (mods are in
red):
	 
	       DESCRIPTION  
	            " This object contains the MTA Admin Status of this
device. 
	              If this object is set to 'true', the MTA is  
	              administratively enabled and the MTA MUST be able to  
	              interact with the PacketCable entities such as CMS,  
	              Provisioning Server, KDC, and other MTAs and MGs on all

	              PacketCable interfaces. 
	              If this object is set to 'false', the MTA is  
	              administratively disabled and the MTA MUST perform the

	              following actions for all endpoints: 
	                  - shutdown all media sessions if present, 
	                  - shutdown NCS signaling by following the Restart
in 
	                  Progress procedures in the PacketCable NCS 
	                  specification. 
	
	 
	              MTA must execute all actions required to enable/disable
the 
	              telephony services for all endpoints immediately upon
receipt of 
	              the SNMP SET operation. 
	 
	              Additionally, the MTA MUST maintain the SNMP Interface

	              For management and also SNMP Key management interface.

	              Also, the MTA MUST NOT continue Kerberized key
management  
	              with CMSes until this object is set to 'true'. 
	              Note: MTAs MUST renew the CMS Kerberos tickets
according  
	              to the PacketCable Security or IPCablecom
Specification."     
	 
	 
	Would this modification address your concern ?
	 
	Eugene.
	 
	
	
________________________________

	From: Wijnen, Bert (Bert) [mailto:[email protected]] 
	Sent: Monday, September 12, 2005 2:03 PM
	To: Eugene Nechamkin; Jean-Francois Mule; Ipcdn (E-mail)
	Cc: Richard Woundy @ Comcast
	Subject: RE: AD Review: draft06 draft-ietf-ipcdn-pktc-mtamib-06.txt (
Comm ent 13)
	
	
	Maybe add some of that explanation to teh DESCRIPTION clause then.
	 
	Bert

		-----Original Message-----
		From: Eugene Nechamkin [mailto:[email protected]]
		Sent: Monday, September 12, 2005 22:19
		To: Wijnen, Bert (Bert); Jean-Francois Mule; Ipcdn (E-mail)
		Cc: Richard Woundy @ Comcast
		Subject: RE: AD Review: draft06
draft-ietf-ipcdn-pktc-mtamib-06.txt ( Comm ent 13)
		
		
		> For example, if I set AdminStatus to "enable", and if the
device
		> seems unable to do so now, then I would still want the
device to
		> try and do it as soon as it can.
		 
		This example tells me where the disconnection might be. Let
me try once more: the PacketCable Provisioning Specification prohibits the
conditions when "the device seems unable to do so now". As soon as the EMTA
receives the SNMP SET with "pktcMtaDevEnabled" being set to true/false, the
device must and is always capable of enabling/disabling the telephony
services on the EMTA. So, there is no scenario which can make "the device to
try and do it as soon as it can" as the device would have executed all
actions required by the corresponding Admin state upon SNMP SET operation.
The conditions of your concern cannot happen in PacketCable environment with
PacketCable compliant EMTA device.
		 
		Please elaborate if this still does no cover your concerns on
this subject.
		 
		Eugene.
		 

________________________________

		From: Wijnen, Bert (Bert) [mailto:[email protected]] 
		Sent: Monday, September 12, 2005 12:53 PM
		To: Eugene Nechamkin; Jean-Francois Mule; Ipcdn (E-mail)
		Cc: Richard Woundy @ Comcast
		Subject: RE: AD Review: draft06
draft-ietf-ipcdn-pktc-mtamib-06.txt ( Comm ent 13)
		
		
		Inline

			-----Original Message-----
			From: Eugene Nechamkin
[mailto:[email protected]]
			Sent: Monday, September 12, 2005 21:24
			To: Wijnen, Bert (Bert); Jean-Francois Mule; Ipcdn
(E-mail)
			Cc: Richard Woundy @ Comcast
			Subject: RE: AD Review: draft06
draft-ietf-ipcdn-pktc-mtamib-06.txt ( Comment 13)
			
			
			
			 
			Because the message is becoming too big I am
separating the responses for each comment for better trackability (also
modifying the title slightly).
			 
			# Comment 13
			 
			
			> - For pktcMtaDevEnabled a better name woul probably
be
			>       pktcMtaDevAdministrativelyEnabled
			>   And how is the!
			  NMS going to see what the real Operational status
is?
			>   Is a AdminStatus and OperStatus (as in IF-MIB)
not more appropriate?
			Proposed resolution: no action
			In general, we agree with the comment and our first
reaction was, yes,
			let's rename it and add an operStatus one.
			But after looking at its definition in more details,
syntax
			(ThruthValue) and after c
			onsidering a new operStatus object, we think
			that:
			        - pktcMtaDevEnabled is both an admin and
operational status
			          object, hence the name is probably ok,
			        - it returns the operational status of the
enable/disable
			          switch when read.
			Our proposal, after reviewing your comment is still
to leave it as-is.
			Is this ok? if not, can you elaborate?
			 
			> bw: The issue is that if you use one single object
for both the admin's desire of the status and
			> at the same time for the actual operational status,
then it gets difficult. 
			> If you issue an SNMP SET to enable, then if you
accept it, then that means that the value
			> is to be changed to enabled. If however the value
then later changes, then I bet that the
			> management station is at least very surprised. It
is not for nothing that the IF-MIB (after a long
			> experience) has chosed to haev a Admin and
Operational status in two !
			 separate objects.
			 
			Let me put some more clarification on the issue which
may address your general concern with the Admin and Operational Status
separation. One of the reasons it would make sense to separate Admin and
Operational status is the potentail delay between the the SNMP SET changing
the Admin Status and actual Operational status of the device including the
case when the device might be not able to set the corresponding Admin status
for some reasons internal to the device. In case of EMTA Enabling/Disabling,
the PacketCable Provisioning spec does not allow such delays and requires the
EMTA to disable/enable the Telephony Services on the EMTA with very minimal
delay (if at all). Hence, from the operational prosective, the separation of
the Admin and Operational Statuses in PacketCable environment does not seem
to be required.
			 
			Hope this clarifies. Please elaborate further
otherwise.
			 
			Jean-Francois and Eugene.
			 
			
			bw: nope it does not clarify (at least not for me).
			I always see a difference between what an operator
wan!
			 ts/intends
			(i.e. AdminStatus) and what the device was actually
able to do
			or!
			  has done (i.e. OperStatus). The timing as to how
fast the 
			device/software has to do it has nothing to do with
it it in my view.
			For example, if I set AdminStatus to "enable", and if
the device!
			 
			seems unable to do so now, then I would still want
the device to
			try and do it as soon as it can.
			 
			
			Bert

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