Ron, Nice to see you participating on this list.
When you proposed your MPLS transport format for 1588, I was very much against it (I freely admit). The reason I was against was that I knew that if we allowed this, then the next step was to demand TC functionality, and no LSR manufacturer I knew was interested in that happening. However, over the last year I have changed my mind. The reason is MPLS-TP. I can see MPLS-TP switches (I am trying to call them a different name from LSR) for special-purpose networks, such as cellular SPs, implementing all sorts of on-path support mechanisms. So it took a long time, but I have finally come around. I think one of the things TICTOC should do is to decide how to best handle this issue. If we decide that the default method (AKA Shahram's method) is optimal, or even "good enough" then we can either leave it at that, or come up with a BCP. If we decide that we need something more scalable, then there are various options, and we need to produce a standard's track document. Y(J)S From: [email protected] [mailto:[email protected]] On Behalf Of Ron Cohen Sent: Tuesday, July 13, 2010 23:49 To: Shahram Davari Cc: [email protected] Subject: Re: [TICTOC] FW: 1588 over MPLS draft Hi Shahram, Its been a while since I read this draft as well.... After reading it again and reading your draft the second time, indeed the issue identification of PTP LSP by intermediate TCs is not addressed in the draft I wrote. This can be solved by extending the PTP FEC signaling defined in the draft between ordinary and boundary clocks to also address signaling between all PTP aware LSRs (i.e. including TCs). The approach described in your draft is also a valid starting point. Best, Ron On Tue, Jul 13, 2010 at 8:03 PM, Shahram Davari <[email protected]<mailto:[email protected]>> wrote: Hi Ron, Hope you are doing fine. Sorry for my comment, I had read it a while back and couldn't remember it precisely. However, let's discuss it: 1) How do the Intermediate LSRs detect that a packet contains PTP? Do you require snooping at line rate? 2) How is a normal Ethernet or ATM or FR PW packet that uses CW is differentiated from a packet carrying PTP? Thanks, Shahram From: Ron Cohen [mailto:[email protected]<mailto:[email protected]>] Sent: Monday, July 12, 2010 10:38 PM To: Shahram Davari Cc: Michel Ouellette; [email protected]<mailto:[email protected]> Subject: Re: [TICTOC] FW: 1588 over MPLS draft Shaharm, SD> I know Ron had a nice draft, but the issue with that draft is that it required Deep packet inspection at line rate for all packets. Also it requires use of CW. Thanks for the compliment, but you probably read someone's else draft. Here it is http://tools.ietf.org/html/draft-ronc-ptp-mpls-00 The draft defines a PTP PW FEC (see section 6), hence it doesn't require any inspection. CW is not used (see figure 5). It does specify a direct PTP over MPLS mapping, without the additional use of Ethernet or IP encapsulation. Best, Ron
_______________________________________________ TICTOC mailing list [email protected] https://www.ietf.org/mailman/listinfo/tictoc
