Re: [PWE3] once upon an erratum

Agree with Stewart.  It would be good to have a draft clearly documenting the gaps and current practices operators use to overcome the challenges.  This will provide a base for solid discussion on whether protocol extensions are required.
When I worked for operators before, I have been working on MEF services over MPLS.  I remember we could do service bundling without considering extension to PWE3 standards.

Raymond Key 
> Date: Wed, 9 May 2012 15:29:07 +0100
> From: stbryant@cisco.com
> To: Alexander.Vainshtein@ecitele.com
> CC: yaakov_s@rad.com; pwe3@ietf.org
> Subject: Re: [PWE3] once upon an erratum
> 
> This sounds like it needs a short RFC explaining the issue and
> if necessary updating RFC4447
> 
> Stewart
> 
> On 09/05/2012 14:48, Alexander Vainshtein wrote:
> > Yaakov,
> > Lots of thanks for a prompt response (somewhat delayed by my persistent errors in the references:-).
> >
> > After looking up again 5603 I tend to agree with you that some of the MEF scenarios are not covered. E.g., the situation when multiple (but not all) 802.1q tags are mapped to the same PW. IMHO it is also not covered by 4447 - there is no way to assure via signaling that the same set of VLANs is mapped to a given single PW at two endpoints of a tagged Ethernet PW.
> >
> > Maybe we should start from explicitly listing various MEF scenarios, mapping some of them to the Ethernet PW parameters (both local and signaled) and coming up first with a list of gaps and then with solutions that would bridge these gaps.
> >
> > I am not sure if there is sufficient interest in this kind of work in the community; were it otherwise, SPs would be crying out loud for these solutions. Or do I miss something?
> >
> > Regards,
> >       Sasha
> >
> >
> >> -----Original Message-----
> >> From: Yaakov Stein [mailto:yaakov_s@rad.com]
> >> Sent: Wednesday, May 09, 2012 4:34 PM
> >> To: Alexander Vainshtein; stbryant@cisco.com
> >> Cc: pwe3@ietf.org
> >> Subject: RE: [PWE3] once upon an erratum
> >>
> >> Sasha,
> >>
> >> Unfortunately, IMO 5603 is not clear, and is possibly not comprehensive.
> >>
> >> 5603 defines 3 cases (and states that all others are out of scope),
> >> that I understand as follows :
> >>
> >> 1) port mode - i.e. All-to-one bundling
> >>
> >> 2) single VLAN - i.e., look at the first VLAN, and then perform one-to-one
> >> bundling
> >>     2.a with tag preservation
> >>     2.b with VLAN swap
> >>     2.c VLAN removal and the PW label identifying the EVC
> >>     2.d with untagged frames passed without modification
> >>     2.e with untagged frames passed after inserting PVID
> >>     2.f with S-tag insertion
> >>
> >> 3) Q-in-Q
> >>     3.a one-to-one with outer VLAN tag used as in 2 above (in fact, I think this
> >> is precisely case 2!)
> >>     3.b one-to-one with both VLAN tags used and the outer tag removed.
> >>
> >> My understanding that all the cases in 2 and 3 are one-to-one
> >> and not "arbitrary bundling" may be incorrect.
> >> I base this understanding on the settings, but perhaps these are just
> >> examples.
> >>
> >> Y(J)S
> >
> > 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.
> >
> >
> 
> 
> -- 
> For corporate legal information go to:
> 
> http://www.cisco.com/web/about/doing_business/legal/cri/index.html
> 
> 
> _______________________________________________
> pwe3 mailing list
> pwe3@ietf.org
> https://www.ietf.org/mailman/listinfo/pwe3