Re: [ANCP] comments on draft-ietf-ancp-mib-an-10

[Moti Morgenstern <[e-mail]>]

Hi,

See inline my comments.

Regards,
Moti

-----Original Message-----
From: [e-mail] [mailto:[e-mail]] On Behalf Of Wojciech Dec (wdec)
Sent: Wednesday, February 13, 2013 5:11 PM
To: [e-mail]
Cc: [e-mail]; [e-mail]
Subject: [ANCP] comments on draft-ietf-ancp-mib-an-10

Hi All,

Here's my review/feedback regarding the draft-ietf-ancp-mib-an-10 draft:

Section 4.
While it may be obvious, it likely would be useful to add that the ANCP Server MIB document is not scoped by this draft. Suggested text could be something like:

"This document specifies a MIB module an ANCP client, such as an Access Node (AN). An ANCP server, such as a NAS, would be typically expected to implement an ANCP server MIB that is not covered by this document."

**** MM: OK ****

Section 5.4 (but applicable to whole document as it pops up throughout) It would be preferable to recast sentences such as  "This table is used to configure ANCP sessions at the AN towards a
      specific NAS"

To:
This table is used to configure ANCP sessions at the ANCP client towards a
      specific ANCP server.

**** MM: OK ****

Section 5.4
Text says: "Each session configured in ancpAnSessionConfigTable has a corresponding row in ancpAnCurrentSessionTable."

As currently defined there seems to be little difference between the ancpAnSessionConfigTable and the ancpAnCurrentSessionTable.
Shouldn't the intent be that only active sessions, makred as such in the ancpAnSessionConfigTable, be represented in the ancpAnCurrentSessionTable?

**** MM: 
The description clause of the ancpAnCurrentSessionTable currently says that "A row in this table is created when the corresponding row in the ancpAnSessionConfigTable is created". I agree that it is too early to expect any status associated with a session that was just created and still isn't fully configured. That is, the row in the status table is meaningful only after the row in the configuration table is activated. So, there is a need to correct the description.
However, I think that the term "active sessions" actually covers sessions that their ancpAnSessionConfigRowStatus object is either set to 'active' or 'notInService'. 
****

Section 7.

AncpSessionCapabilities

The capabilities list does not reflect the capabilities we have in the defined in other ANCP specs.
Missing are:
- Multicast
(http://tools.ietf.org/html/draft-ietf-ancp-mc-extensions-08#section-6)
- PON (TBD)

**** MM: 
There are plenty of capabilities in ANCP but I didn't manage to find any "list". So, probably 'multicast' and 'PON' should be added. I assume we need also to add 'Line Connectivity Testing'. But, I also assume that those three are not the only missing capabilities.
****

The names of the capabilities shown should be aligned to what we have in:
http://tools.ietf.org/html/rfc6320#section-10.7

**** MM: OK ****

ancpAnSessionConfigNasIpAddressType and ancpAnCurrentSessionAnIpAddressType

Shouldn't the syntax also detail the allowed values for say InetAddressType? Eg
SYNTAX     InetAddressType { ipv4(1), ipv6(2) }

**** MM: 
I think that the following two actions are required:
1) Add a sentence in the description clause of each object saying: "While other types of addresses are defined in the InetAddressType textual convention, this object is limited to IPv4
       and IPv6 addresses." 
2) Put a limit in the MODULE-COMPLIANCE statement, e.g.

    OBJECT ancpAnSessionConfigNasIpAddressType
    SYNTAX  InetAddressType { unknown(0), ipv4(1), ipv6(2) }
    DESCRIPTION
       "An implementation is only required to support IPv4 and IPv6 addresses."
****

ancpAnSessionConfigNasIpAddress and ancpAnCurrentSessionAnIpAddress

Shouldn't the syntax also indicate the size? Eg SYNTAX InetAddress ( SIZE (4|16) )

**** MM: 
I think that the following two actions are required:
1) Indicate in the description clause of each object that the object is limited to IPv4 and IPv6 addresses. 
2) Put a limit in the MODULE-COMPLIANCE statement, e.g.

    OBJECT ancpAnSessionConfigNasIpAddress
    SYNTAX  InetAddress (SIZE(0|4|16))
    DESCRIPTION
       "An implementation is only required to support IPv4 and globally unique IPv6 addresses."
****

Why isn't there an ancpANSessionConfigIPAddress* ? If the assumption is that this address becomes derived only from the IF-MIB, and the assignment of interfaces to the ANCP session, then it would be helpful to explain that to the reader.

Regards,
Woj..




_______________________________________________
ANCP mailing list
[e-mail]
https://www.ietf.org/mailman/listinfo/ancp


This e-mail message is intended for the recipient only and contains information which is CONFIDENTIAL and which may be proprietary to ECI Telecom. If you have received this transmission in error, please inform us by e-mail, phone or fax, and then delete the original and all copies thereof.