Re: [mpls] Alexey Melnikov's Discuss on draft-ietf-mpls-rfc3107bis-02: (with DISCUSS and COMMENT)

["Acee Lindem (acee)" <[e-mail]>]

Hi Alexey, 

On 8/17/17, 9:14 AM, "Alexey Melnikov" <[e-mail]> wrote:

>Hi Acee,
>
>On Thu, Aug 17, 2017, at 02:11 PM, Acee Lindem (acee) wrote:
>> Hi Alexey,
>> 
>> See one comment below.
>> 
>> On 8/17/17, 8:05 AM, "mpls on behalf of Alexey Melnikov"
>> <[e-mail] on behalf of [e-mail]> wrote:
>> 
>> >Hi Eric,
>> >Sorry for the slow response.
>> >
>> >On Tue, Aug 1, 2017, at 09:59 PM, Eric C Rosen wrote:
>> >> 
>> >> > DISCUSS:
>> >> > 
>>----------------------------------------------------------------------
>> >> >
>> >> > I would like to discuss one issue before recommending approval of
>> >>this document:
>> >> >
>> >> > In Section 2.1:
>> >> >
>> >> >     The value field of the Multiple Labels Capability (shown in
>> >>Figure 1)
>> >> >     consists of one or more triples, where each triple consists of
>> >>four
>> >> >     octets.  The first two octets of a triple specify an AFI value,
>> >>the
>> >> >     third octet specifies a SAFI value, and the fourth specifies a
>> >>Count.
>> >> >     If one of the triples is <AFI,SAFI,Count>, the Count is the
>> >>maximum
>> >> >     number of labels that the BGP speaker sending the Capability
>>can
>> >> >     process in a received UPDATE of the specified AFI/SAFI.
>> >> >
>> >> > I think lack of recommendations on the minimal supported Count
>>value
>> >>will
>> >> > result in lack of interoperability. What are the common Count
>>values
>> >>used by
>> >> > implementations?
>> >> 
>> >> An implementation is not required to support more than one label in
>>the
>> >> NLRI.  In that case, of course the Multiple Labels Capability is not
>> >> supposed to be used.  So if the Capability is used, the minimum value
>> >>is 
>> >> 2.   It looks like the document does not state that explicitly, but
>>it
>> >> certainly should.  I will fix that.
>> >> 
>> >> Thus the literal answer to your question is that the Multiple Labels
>> >> Capability MUST NOT contain a count less than 2.
>> >> 
>> >> Typically when SAFI-4 or SAFI-128 routes are used, only one label is
>> >> included.  If an operator has a scenario in which multiple labels are
>> >> needed, it is necessary for the operator to ensure that all his
>>vendors
>> >> can support however many labels he needs.  That's not really
>>something
>> >> that the standard can address.
>> >
>> >The standard can specify minimum requirement (and this has been done in
>> >the past). For example you can say that all implementations must
>>support
>> >4 (or whatever).
>> >
>> >> If the operator needs 3 labels in the
>> >> NLRI, it won't help much if some of his boxes can only support 2.
>> >
>> >Exactly my point. Without the minimum you still need to have
>>out-of-band
>> >selection of implementations that support your minimum requirements.
>> >
>> >> ---------------------------------------------------------------------
>> >> > COMMENT:
>> >> > 
>>----------------------------------------------------------------------
>> >> >
>> >> > In Section 2.3:
>> >> >
>> >> >        0                   1                   2
>>  3
>> >> >        0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9
>>0 1
>> >> >       +-+-+-+-+-+-+-+-+
>> >> >       |    Length     |
>> >> >       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
>> >> >       |                 Label                 |Rsrv |S~
>> >> >       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
>> >> >       ~                 Label                 |Rsrv |S|
>> >> >       
>> >>+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
>> >> >       |                          Prefix
>> >>~
>> >> >       ~    
>> >>|
>> >> >       
>> >>+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
>> >> >
>> >> >                      Figure 3: NLRI With Multiple Labels
>> >> >
>> >> >     - Length:
>> >> >
>> >> >        The Length field consists of a single octet.  It specifies
>>the
>> >> >        length in bits of the remainder of the NLRI field.
>> >> >
>> >> > I would like to double check that my math is correct. With SAFI=128
>> >>and AFI=2,
>> >> > assuming the prefix length of 192 bits, this will leave space for:
>> >> >
>> >> >   (255-192)/24 = 2.625. So this configuration only allows for 2
>> >>labels to be included, right?
>> >> >
>> >> Your arithmetic is accurate ;-)
>> >> 
>> >> If you are implying that this is not necessarily the best mechanism
>>for
>> >> associating an arbitrarily long label stack with a prefix, I wouldn't
>> >> disagree.
>> >
>> >Yes, I was just surprised that this has such a small limit. I wanted to
>> >make sure that this is by design.
>> 
>> Note that there are more complex BGP encodings in IDR WG documents that
>> support larger label stacks.
>> 
>> https://www.ietf.org/id/draft-ietf-idr-segment-routing-te-policy-00.txt
>> https://www.ietf.org/id/draft-ietf-idr-tunnel-encaps-07.txt
>> 
>> This much-needed document simply clears up the ambiguities of
>>advertising
>> multiple labels with the RFC 3107 NLRI encodings.
>
>So just to double check: you are saying that this is cleaning up a
>legacy specification and there are better/more modern specifications
>that can do the same thing?

It is more complicated than that. There are other considerations with
using tunnel attributes and still others with SR policies. RFC 3107 NIS is
needed. 

Thanks,
Acee 
>