Re: [Bier] WG adoption call for draft-chen-bier-frr-02

[[e-mail]]

Hi Tony,





When primary NBR has failed, the following two schemes have the same logic, both of them limit the processing socpe to the failure of specific primary NBR.

1) Within the scope of specific FRR-BIFT for the primary NBR, when packet is steered to an FRR-BIFT entry, the bitstring contained in the outgoing copy is the result of "original bitstring of the received packet" & "FBM".

2) A single BIFT entry with primary NBR and backup path, when packet is steered to backup path, the bitstring contained in the outgoing copy is the result of "original bitstring of the received packet" & "primary FBM" & "backup FBM".

For example, BIFT entry 1 has primary NBR X, BIFT entry 2 has primary NBR Y, but they have the same backup path. When NBR X has failed, BIFT entry 1 need switch to backup path, but BIFT entry 2 need not.




If the above two schemes are equivalent,  I think the simple one is the better one. There is no need to maintain so many FRR-BIFT tables as draft-chen described.




Regards,

PSF










原始邮件



发件人：TonyPrzygienda
收件人：Gengxuesong (Geng Xuesong);
抄送人：Jeffrey (Zhaohui) Zhang;Aijun Wang;BIER WG;张征00007940;BIER WG Chairs;
日 期 ：2021年03月23日 04:36
主 题 ：Re: [Bier] WG adoption call for draft-chen-bier-frr-02







On Fri, Mar 19, 2021 at 11:42 AM Tony Przygienda <[e-mail]> wrote:




I would prefer a single document on bier-frr as joined Menth and Xuesong draft with the single BIFT mode & possibly multiple (again, modulo technical comments on it right now) but it's up to the authros to sit down and od the work and drink some beer together if they choose. 






I stand corrected of course, it was Huaimo's draft 


-- tony