Monday, August 29, 2022< ^ >
Meetecho has set the subject to: IDR IETF-113
Room Configuration
Room Occupants

[14:04:14] <zulipbot> (Jie Dong) here is the link to the minutes:
[14:12:09] <zulipbot> (Jeffrey Haas) Linda, you are faint and sound like you're on an open air mic.
[14:13:48] <zulipbot> (Jeffrey Haas) Randy, did you have a question?
[14:20:42] <zulipbot> (Natrajan Venkataraman) I am having some audio issues. Not able to hear. Will try resolving
[14:25:27] <zulipbot> (Natrajan Venkataraman) Resolved
[14:30:03] <zulipbot> (Srihari Sangli) lost audio
[14:30:04] <zulipbot> (John Scudder) Is anyone receiving audio?
[14:30:16] <zulipbot> (Tony Li) Nope
[14:30:17] <zulipbot> (Keyur Patel) lost audio
[14:33:51] <zulipbot> (Srihari Sangli) Comments on how the BGP bestpath will churn as a result of these additional sub-TLV i.e. dynamic capacity or load  ?
[14:45:25] <zulipbot> (Ketan Talaulikar) I noticed that one of the two drafts is posted as WG standards track. Seems like a mistake then?
[14:45:38] <zulipbot> (Jeffrey Haas) Those will be getting corrected, Ketan.
[14:45:57] <zulipbot> (Jeffrey Haas) A github has been setup for both drafts and we'll work through the various issues there and the mailing list
[14:46:42] <zulipbot> (Ketan Talaulikar) Thanks for confirming Jeff. Hopefully ASAP since it has the potential to cause undue confusion
[15:20:51] <zulipbot> (Jie Dong) I think Haibo means he plans to reply to Kaliraj's other comments by email
[15:33:00] <zulipbot> (Linda Dunbar) For the Edge Service Metadata draft, to prevent churn, Section 7 discusses the  Minimum Interval for Metrics Change Advertisement
[15:45:05] <zulipbot> (Aijun Wang) For the Mapping nomenclature:
should the “CAR-mapped-CT route(When a BGP-CAR route carries BGP-CT original intent,)” be changed to "CT-mapped CAR route"

and the “CT-mapped-CAR route( When a BGP-CT route carries BGP-CAR original intent)” be changed to "CAR-mapped CT route"

I think it is more acceptable as the concept of  "IPv4-mapped IPv6 address"
[15:46:24] <zulipbot> (Robert Raszuk) @Jeff - does this work only between domains ? So what if I have in one domain 500 routes which speak only BGP-CAR and 500 routers which speak only BGP-CT ?
[15:54:44] <zulipbot> (Kaliraj Vairavakkalai) - kiss principle : inter-as option a, b.
   - slide 5 effective color: ext-color-community?
   - slide 15.3: resolution procedures similar?
   - slide 11-14: CTOI not needed?
   - slide 11.5: delete TC-RT - dangerous.
[15:57:22] <zulipbot> (Robert Raszuk) Also as I asked on the list two more points:
a) how do you handle incremental additional functionality additions to either CAR or CT once interop is running say by third vendor's  running code on RRs or ASBRs
b) when there was discussion there was a lot of indications for single solution. Would we still need to interop if we have a single solution adopted ?
[15:59:20] <zulipbot> (Susan Hares) Robert - type b) please clarify if you consider Jeff's proposal a single solution.
[15:59:34] <zulipbot> (Srihari Sangli) @Robert - Option (b) is only needed if we have deployments of CAR or CT before the third soln is available. I guess operators can speak on their liking
[15:59:47] <zulipbot> (Robert Raszuk) Nope by single solution I mean either BGP-CAR or BGP-CT
[16:00:34] <zulipbot> (Susan Hares) Robert - then your question on option (b) is out of scope for this interim.
[16:02:08] <zulipbot> (Srihari Sangli) nope Jeff
[16:05:45] <zulipbot> (Keyur Patel) no comments
[16:05:58] <zulipbot> (Tony Li) Thank you!
[16:06:12] <zulipbot> (Jie Dong) thanks everyone
[16:06:25] <zulipbot> (Kaliraj Vairavakkalai) Thank you