Re: [mpls] Question on draft-lm-mpls-sfc-path-verification

[[e-mail]]

Hi Loa,

Thanks for your questions and suggestions.

Please see in line with [Yao].







原始邮件



发件人：LoaAndersson <[e-mail]>
收件人：刘尧00165286;
抄送人：[e-mail] <[e-mail]>;[e-mail] <[e-mail]>;[e-mail] <[e-mail]>;
日 期 ：2020年09月18日 11:16
主 题 ：Re: Question on draft-lm-mpls-sfc-path-verification


Yao,

So most is sorted out :).

Some small issues.

On 17/09/2020 18:47, [e-mail] wrote:
> 
> The SFC Basic Unit FEC Sub-TLV maybe included in a SFC Validation TLV
> carried both in a MPLS echo request and a MPLS echo reply.
> 
>  Currently 
"currently means that it is defined in section 3.3 of 
draft-lm-mpls-sfc-path-verification, right. I find the definition is 3.3 
rather weak, more detals is needed. e.g. that the sub-tlv is defined for 
FEC Stack TLV (TLV #1).




[Yao] Got it. The details of the definition and processing procedure will be added to the section.




the SFC Basic Unit FEC Sub-TLV is denfined as a sub-TLV

> of the Target FEC Stack TLV,  carried in a MPLS echo request. Its usage 
> is similar to other sub-TLVs of the Target FEC Stack. 
ok- fine, but the sub-TLVs are shared between TLV 1, 16, and 21. Is that 
true for the SFC Basic Unit FEC Sub-TLV?




[Yao] If the SFC is bi-directional, the SFC Basic Unit FEC Sub-TLV can be carried in TLV 16 for reverse SFP connectivity verification or in TLV 21 to carry the specified return SFP info to be used by the echo reply.  There are more details needed in this section as you said. I'll make it clearer.




The information
> carried in the Basic Unit FEC Sub-TLV is obtained from the control 
> plane.  After receiving a MPLS echo request with the Target FEC Stack 
> TLV and the SFC Basic Unit FEC Sub-TLV included, an SFF may enter the 
> FEC validation procedure to check whether the information is the same 
> between the control place and the local data plane. If the validation is 
> not passed, the SFF will generate an MPLS echo reply with an error code 
> as defined in RFC8029.
OK - and you need to point out the range from which the sub-tlv should 
be allocated.




[Yao] OK, I will add the paragraph about the allocation in the next version and I believe that the sub-TLV should be allocated from range 0-16383, which is standards action.




/Loa


PS

Plz note that we are doing updates to the LSP Ping registry, so far I 
have not seen anything that directly have an impact on this draft.
See - draft-ietf-mpls-lsp-ping-registries-update-.

[Yao] Thanks for reminding~ I'll go through the draft to learn more about it. 

> 
> 

-- 

Loa Andersson                        email: [e-mail]
Senior MPLS Expert                          [e-mail]
Bronze Dragon Consulting             phone: [phone]