James, I am adding PALS WG to the list of addressees because, AFAIK, the PW CW is defined in this WG.
I think that the really observed problem with incorrect ECMP behavior exists, but it is different from your description in your earlier email: Any LSR on the path between ingress and egress LER is free to look beyond the MPLS label stack and misinterpret the 0x00 0x00 at the start of a control-word as a valid MAC that starts 00:00:XX:XX:XX:XX and try to hash on Ethernet headers starting directly after the MPLS label stack. I have not seen (or heard about) such behavior in any deployed networks. However, I am aware of some modern forwarding chipsets that (correctly) treat the ‘0000’ in the first nibble of the payload of a labeled packet (i.e., immediately following the bottom of the label stack) as the indication of a 32-bit PW control word but (incorrectly), consider this as a CW of an Ethernet PW (as if no other PWs exist!) and try to hash on the presumed MAC addresses, Ethertype etc. Such behavior is really deadly for, say TDM PWs that, AFAIK, are still widely deployed in many places. I wonder if IETF (or, specifically, the PALS WG) can do anything about that. Regards, Sasha Office: +972-39266302 Cell: +972-549266302 Email: [email protected] -----Original Message----- From: BESS [mailto:[email protected]] On Behalf Of James Bensley Sent: Tuesday, October 23, 2018 11:06 AM To: [email protected] Subject: [bess] CW in EVPN: Was Signaling Control Word in EVPN Hi All, I've removed all the individual contacts as per mod request and changed the subject as this conversation was diverging away from the OPs original query (sorry if any of that was my fault). On Tue, 9 Oct 2018 at 15:54, Andrew G. Malis <[email protected]<mailto:[email protected]>> wrote: > > James, > > Agreed. We touched on that in section 7 of draft-ietf-pals-ethernet-cw, where > we advised operators that enabling post-CW DPI for ECMP calculations could > cause misordering. Hi Andy, Perhaps I wasn't clear enough - and LSR (not the ingress/egress LER) even trying to detect CW is a source of problems within the network, this is why I was querying the use of the PWMCW. On Tue, 9 Oct 2018 at 16:44, Yutianpeng (Tim) <[email protected]<mailto:[email protected]>> wrote: > > Personally I support having control word capability which is more common. It > is said in RFC 8214: > > “It is recommended that the control word be included in the absence of an > entropy label” > > So I think we can still have EL capability on EVPN VPWS if I understand > correct. Only if the existing "BGP Path Attributes" does not work well. > > I support adapt CW capability into EVPN also as EVPN itself is facing same > challenge with EVPN VPWS. > > By the way, rfc4385 mentioned PW controlling ECMP. The CW doesn't control ECMP - this is my gripe here. It's one thing to recommend a technology that not all devices support (EL/ELI or FAT) but which fully fixes the problem if it supported, it's another thing to recommend a technology that doesn't fully fix the problem if it is supported at all. My concern is that by continuing to suggest a partial solution (PWMCW) we help to keep that partial solution in circulation when surely the WG should look to deprecate it for a solution that fully fixes the problem? Cheers, James. _______________________________________________ BESS mailing list [email protected]<mailto:[email protected]> https://www.ietf.org/mailman/listinfo/bess ___________________________________________________________________________ This e-mail message is intended for the recipient only and contains information which is CONFIDENTIAL and which may be proprietary to ECI Telecom. If you have received this transmission in error, please inform us by e-mail, phone or fax, and then delete the original and all copies thereof. ___________________________________________________________________________
_______________________________________________ BESS mailing list [email protected] https://www.ietf.org/mailman/listinfo/bess
