Re: [CCAMP] 2nd WG last call on draft-ietf-ccamp-gmpls-ted-mib

[Acee Lindem <[e-mail]>]

Hi Masanori,


On Apr 30, 2011, at 8:50 PM, Masanori Miyazawa wrote:

> Hi, Ben
> 
> Thank you for your comments.
> I have my answer for your comments inline below.
> If you have any comments, please let us know.
> 
> With best Regards,
> Masanori
> 
>> *	The tedTable structure doesn't make it easy for network
>> administrators to query just the TED entries that are learned from, say,
>> OSPFv2 (rather than ISIS).  An easy fix for this would be to make the
> primary
>> index field the information source (OSPFv2, ISIS etc).  This would also
>> make the interpretation of the syntax of the remaining index fields
>> straightforward for MIB agents.
> 
> To simplify the index field, we define a new primary index (e.g. tedIndex)
> and use it as index in the tedTable as below. Then, I think you have
> accessibility to TE link information.  


What is the definition of the this index? You don't want to leave it up to the implementation as this will lead to different semantics.

Also, you should allow 0 in the range for Area IDs as 0 is certainly a valid OSPF area ID. 

  
   TedAreaIdTC ::= TEXTUAL-CONVENTION 
          DISPLAY-HINT "d" 
          STATUS       current 
          DESCRIPTION 
             "The area identifier of the IGP. If OSPF is used to 
   advertise LSA, this represents an ospfArea. If ISIS is used, this 
   represents an area address." 
          SYNTAX       Unsigned32 (1..4294967295) 

Should be: 
          SYNTAX       Unsigned32 (0..4294967295)  


Thanks,
Acee 


> 
> tedEntry OBJECT-TYPE
>    SYNTAX       TedEntry
>    MAX-ACCESS   not-accessible
>    STATUS       current
>    DESCRIPTION
>    "This entry contains TED information commonly utilized in both MPLS and
> GMPLS."
>   INDEX { tedIndex }
> ::= { tedTable 1 }
> 
> TedEntry ::= SEQUENCE {
>    tedIndex                     Unsigned32,
>    tedAreaId                    TedAreaIdTC,
>    tedRouterId                  TedRouterIdTC,
>    tedLinkStateId               TedLinkStateIdTC,
>    tedLinkInformationSource     INTEGER,
>    tedLinkInformationData       RowPointer,
>    tedLinkState                 INTEGER,
>    tedLinkType                  INTEGER,
>    tedRouterIdAddrType          InetAddressType,
>    tedRouterIdAddr              InetAddress,
>    tedLinkIdAddrType            InetAddressType,
>    tedLinkIdAddr                InetAddress,
>    tedMetric                    Integer32,
>    tedMaxBandwidth              Unsigned32,
>    tedMaxReservableBandwidth    Unsigned32,
>    tedUnreservedBandwidthPri0   Unsigned32,
>    tedUnreservedBandwidthPri1   Unsigned32,
>    tedUnreservedBandwidthPri2   Unsigned32,
>    tedUnreservedBandwidthPri3   Unsigned32,
>    tedUnreservedBandwidthPri4   Unsigned32,
>    tedUnreservedBandwidthPri5   Unsigned32,
>    tedUnreservedBandwidthPri6   Unsigned32,
>    tedUnreservedBandwidthPri7   Unsigned32,
>    tedAdministrativeGroup       Integer32,
>    tedLocalId                   Integer32,
>    tedRemoteId                  Integer32,
>    tedLinkProtectionType        BITS
>    }
> 
> 
>> *	The tedLocalIfAddrTable and tedRemoteIfAddrTable both use an
>> arbitrary index to distinguish between the different IP addresses on a
> given
>> link.  Is there a reason why we can't just use the address type and
> address
>> as index fields instead?  That would guarantee, for a given set of IP
>> addresses, the ordering of rows in the table is entirely predictable.
> 
> 
> As you know, note that tedLocalIfAddrTable and tedRemoteIfAddrTable are
> independently defined, because the Interface IP Address sub-TLV may appear
> more than once within the same Link-TLV. Therefore, if the address type and
> the address are used as index, another arbitrary index, such as tedIndex,
> would be also needed to represent the relationship with tedTable as below,
> 
> tedLocalIfAddrTable OBJECT-TYPE
>    SYNTAX       SEQUENCE OF TedLocalIfAddrEntry
>    MAX-ACCESS   not-accessible
>    STATUS       current
>    DESCRIPTION
>      "This table contains the IP address information of a local TE link."
> ::= { tedObjects 2 }
> 
> tedLocalIfAddrEntry OBJECT-TYPE
>    SYNTAX       TedLocalIfAddrEntry
>    MAX-ACCESS   not-accessible
>    STATUS       current
>    DESCRIPTION
>      "This entry contains the IP address information of the local TE link."
>    INDEX { tedIndex, tedLocalifAddr }
> ::= { tedLocalIfAddrTable 1 }
> 
> TedLocalIfAddrEntry ::= SEQUENCE {
>    tedLocalIfAddrType    InetAddressType,
>    tedLocalIfAddr        InetAddress
>    }
> 
> tedLocalIfAddrType OBJECT-TYPE
>    SYNTAX       InetAddressType
>    MAX-ACCESS   read-only
>    STATUS       current
>    DESCRIPTION
> "This object indicates the address type of the local TE link. Only values
> unknown(0), ipv4(1) or ipv6(2) have to be supported." 
> ::= { tedLocalIfAddrEntry 2 }
> 
>> -----Original Message-----
>> From: [e-mail] [mailto:[e-mail]] On Behalf Of
>> Ben Wright
>> Sent: Thursday, April 21, 2011 6:32 PM
>> To: CCAMP
>> Subject: Re: [CCAMP] 2nd WG last call on draft-ietf-ccamp-gmpls-ted-mib
>> 
>> 
>> Hi Masanori et al,
>> 
>> I have a few minor comments on draft-ietf-ccamp-gmpls-ted-mib-08.
>> 
>> * 	I agree with Wenhu's comments on the OSPF mailing list - we need
>> to update the Textual Conventions to take into account ISIS syntax.
>> 
>> *	The tedTable structure doesn't make it easy for network
>> administrators to query just the TED entries that are learned from, say,
>> OSPFv2 (rather than ISIS).  An easy fix for this would be to make the
> primary
>> index field the information source (OSPFv2, ISIS etc).  This would also
>> make the interpretation of the syntax of the remaining index fields
>> straightforward for MIB agents.
>> 
>> *	The tedLocalIfAddrTable and tedRemoteIfAddrTable both use an
>> arbitrary index to distinguish between the different IP addresses on a
> given
>> link.  Is there a reason why we can't just use the address type and
> address
>> as index fields instead?  That would guarantee, for a given set of IP
>> addresses, the ordering of rows in the table is entirely predictable.
>> 
>> Thanks,
>> 
>> Ben
>> -----Original Message-----
>> From: [e-mail] [mailto:[e-mail]] On Behalf Of
>> labn - Lou Berger
>> Sent: 18 April 2011 20:19
>> To: CCAMP
>> Subject: [CCAMP] 2nd WG last call on draft-ietf-ccamp-gmpls-ted-mib
>> 
>> This mail begins a 2nd WG last call on:
>> 
>> http://tools.ietf.org/html/draft-ietf-ccamp-gmpls-ted-mib-08
>> 
>> The draft has been updated after the earlier working group primarily
>> based on MIB Dr. review and discussion on the ccamp list.
>> 
>> This working group last call ends on May 2nd. This LC will be announced
>> on the MPLS, OSPF, and ISIS WG lists.  Please send comments to
>> the CCAMP mailing list.
>> 
>> Lou (and Deborah)
>> _______________________________________________
>> CCAMP mailing list
>> [e-mail]
>> https://www.ietf.org/mailman/listinfo/ccamp
>> _______________________________________________
>> CCAMP mailing list
>> [e-mail]
>> https://www.ietf.org/mailman/listinfo/ccamp
> 
> _______________________________________________
> CCAMP mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/ccamp