Also, as in the private communication, I don’t really see any mac flip-flop issue as we are considering the same ESI. It is "NOT" equivalent to the mac-mobility.
Thanks, --Satya From: "[email protected]<mailto:[email protected]>" <[email protected]<mailto:[email protected]>> Date: Tuesday, April 7, 2015 6:53 AM To: Haoweiguo <[email protected]<mailto:[email protected]>>, "Patrice Brissette (pbrisset)" <[email protected]<mailto:[email protected]>> Cc: "Ali Sajassi (sajassi)" <[email protected]<mailto:[email protected]>>, "[email protected]<mailto:[email protected]>" <[email protected]<mailto:[email protected]>> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm Weiguo, Snipped, comments inline. Yours Irrespectively, John From: Haoweiguo [mailto:[email protected]] Sent: Monday, April 06, 2015 10:03 PM To: John E Drake; Patrice Brissette (pbrisset) Cc: Ali Sajassi (sajassi); [email protected]<mailto:[email protected]> Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm Hi John, Your summary is good. Pls see my further reply inline with [weiguo]. Thanks, weiguo ________________________________ From: John E Drake [[email protected]<mailto:[email protected]>] Sent: Friday, April 03, 2015 21:38 To: Haoweiguo; Patrice Brissette (pbrisset) Cc: Ali Sajassi (sajassi); [email protected]<mailto:[email protected]> Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm Weiguo, I have been marginally paying attention to this email thread and I thought I should update the points we have established. In addition to the points listed in my email, below, I think we have established that; 1) Your proposal is even more broken than I I previously indicated because you have no way to identify to which EVI your DF election handshake applies [weiguo]: Firstly we had better reach consensus on the problem(Loop and MAC flip-flop in transient peirod for MHN) that we are trying to resolve, then we can search for the suitable solution. The current proposed solution is to use A-D route for handshaking, the A-D route can identify EVI. [JD] I don’t think there is a problem to be solved, so the first order of business is for you to write a draft clearly articulating the problems you perceive and their ramifications. Please keep in mind that single-active, and more specifically single-active MHN, is not where I think we should focus our resources. DCI, for example, uses all-active. Further, as I had previously pointed out, we can’t use an EVPN route that goes to all PEs so the Per EVI Ethernet AD route can’t be used.
_______________________________________________ BESS mailing list [email protected] https://www.ietf.org/mailman/listinfo/bess
