Re: [yang-doctors] draft-ietf-isis-sr-yang and draft-ietf-ospf-sr-yang

[Reshad Rahman <[e-mail]>]

 Hi Acee,
Understood. I do realize that there can be good reasons for differences between the 2 documents.
Regards,Reshad. 
    On Tuesday, November 14, 2023, 01:22:23 PM EST, Acee Lindem <[e-mail]> wrote:  
 
 Hi Reshad, 
Note that the SR encodings contain a lot of the same information but are different in the two protocols. It wouldn’t be feasible to use common groupings as it is more importation to be consistent with the data blocks that we are augmenting than the SR extensions in the other protocol. If the IGPs were exactly the same, there would only be one 😎 


On Nov 14, 2023, at 12:17, Reshad Rahman <[e-mail]> wrote:
 Hi Acee,
Couple of other differences (I didn't dig to see whether they are justified):- Naming discrepancies e.g. TLV suffix is used more in OSPF (local-blocks v/s local-blocks-tlv)

The LSDB models for RFC 9129 (OSPF) and RFC 9130 (IS-IS) are somewhat different. It is more important to be consistent with the base models than the other protocol. 


- No global blocks in ISIS

IS-IS has global blocks. 
augment /rt:routing/rt:control-plane-protocols
          /rt:control-plane-protocol/isis:isis/isis:database
          /isis:levels/isis:lsp/isis:router-capabilities:
    +--ro sr-capability
    |  +--ro sr-capability
    |  |  +--ro sr-capability-bits*   identityref
    |  +--ro global-blocks
    |     +--ro global-block* []
    |        +--ro range-size?    uint32
    |        +--ro sid-sub-tlv
    |           +--ro sid?   uint32
    +--ro sr-algorithms
    |  +--ro sr-algorithm*   uint8
    +--ro local-blocks
    |  +--ro local-block* []
    |     +--ro range-size?    uint32
    |     +--ro sid-sub-tlv
    |        +--ro sid?   uint32
    +--ro srms-preference
       +--ro preference?   uint8

- No capabilities in OSPF

We have augmentations for capabilities. 
Refer to https://datatracker.ietf.org/doc/html/rfc8665#name-segment-routing-capabilities
Thanks, Acee




Regards,Reshad.   

    On Tuesday, November 14, 2023, 10:11:02 AM EST, Acee Lindem <[e-mail]> wrote:  
 
 Thanks Reshad - are there any other notable discrepancies? 

Thanks,
Acee

> On Nov 14, 2023, at 9:55 AM, Reshad Rahman <reshad=[e-mail]> wrote:
> 
> Hi,
> 
> My suggestion is that authors of these 2 documents spend some time together to try to align the 2 documents. After that we can do YD review.
> 
> Regards,
> Reshad.
> 
> On Wednesday, November 1, 2023, 10:58:56 AM EDT, Reshad Rahman <[e-mail]> wrote: 
> 
> 
> Hi,
> 
> Background: those 2 documents have just been assigned YD review, I am reviewing OSPF and Jan is reviewing ISIS.
> 
> Was an effort made to keep those 2 documents aligned/in-sync where possible/desirable? My expectation is that the SR specifics would be near-identical in the 2 documents. e.g. shouldn't the capabilities for the 2 protocols be very similar.
> Here are some differences which don't seem justified:
> - sr-algorithm in ISIS is a uint8 and in OSPF is an identityref
> - range-size is a uint32 in ISIS and is a uint24 in OSPF
> 
> 
> augment /rt:routing/rt:control-plane-protocols
> /rt:control-plane-protocol/isis:isis/isis:database
> /isis:levels/isis:lsp/isis:router-capabilities:
> +--ro sr-capability
> | +--ro sr-capability
> | | +--ro sr-capability-bits* identityref
> | +--ro global-blocks
> | +--ro global-block* []
> | +--ro range-size? uint32
> | +--ro sid-sub-tlv
> | +--ro sid? uint32
> +--ro sr-algorithms
> | +--ro sr-algorithm* uint8
> +--ro local-blocks
> | +--ro local-block* []
> | +--ro range-size? uint32
> | +--ro sid-sub-tlv
> | +--ro sid? uint32
> +--ro srms-preference
> +--ro preference? uint8
> 
> augment /rt:routing/rt:control-plane-protocols
> /rt:control-plane-protocol/ospf:ospf/ospf:areas/ospf:area
> /ospf:interfaces/ospf:interface/ospf:database
> /ospf:link-scope-lsa-type/ospf:link-scope-lsas
> /ospf:link-scope-lsa/ospf:version/ospf:ospfv2/ospf:ospfv2
> /ospf:body/ospf:opaque/ospf:ri-opaque:
> +--ro sr-algorithm-tlv
> | +--ro sr-algorithm* identityref
> +--ro sid-range-tlvs
> | +--ro sid-range-tlv* []
> | +--ro range-size? rt-types:uint24
> | +--ro sid-sub-tlv
> | +--ro sid? uint32
> +--ro local-block-tlvs
> | +--ro local-block-tlv* []
> | +--ro range-size? rt-types:uint24
> | +--ro sid-sub-tlv
> | +--ro sid? uint32
> +--ro srms-preference-tlv
> +--ro preference? uint8
> 
> 
> 
> Disclaimer: I don't follow LSR...
> 
> Regards,
> Reshad.
> _______________________________________________
> yang-doctors mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/yang-doctors