Re: [PWE3] GAL above and below the PW label

[Curtis Villamizar <[e-mail]>]

In message <AANLkTi=fuSg1iBy+jz0SvsfNpGrd-=32p+KMV7Q=[e-mail]>
Sriganesh Kini writes:
>  
> See inline
>  
> On Thu, Mar 31, 2011 at 4:40 AM, Curtis Villamizar <[e-mail]> wrote:
> >
> > Sriganesh, Sasha, Loa, Dave,
> >
> > Would you please not top post.  See inline.
> >
> > Curtis
> >
> > In message <[e-mail]>
> > Sriganesh Kini writes:
> >>
> >> Sasha, that issue with VCCV Type-3 (without CW) is not addressed by
> >> the draft. Practically, is that a major issue ? Especially setting TTL
> >> > 1 for SS PW is possible I guess but would be some kind of serious
> >> operational oversight. Besides such a packet would fail some check
> >> along the path.
> >
> > End to end OAM should set TTL>1 and not use VCCV Type-3.  The T-PE may
> > not have reliable information to set TTL, so in MS-PW it might be
> > worth discouraging VCCV Type-3 except for use as a traceroute
> > capability or to address S-PEs by node distance along the path.
>  
> If there is indeed a concern of TTL exceeding number of segments (with
> Type-3 and no CW), the same concern would apply when tracing. So
> tracing would be impossible.

traceroute stops when it hits the egress.

> >> Also, what I heard at the mic regarding impl results is that it
> >> doesn't make sense to deprecate what is already deployed.
> >
> > Yes.  But VCCV Type-3 with MS-PW is not already deployed.
>  
> I am not aware of the survey addressing this.

AFAIK Type-1 is being used, but I could be wrong.  Maybe someone who
has implemented and knows of a deployment or someone who has deployed
can reply so we don't need another survey.

> >> Note that CC Type 1 prevents ECMP (when FAT-PW is not available).
> >> Using GAL below PW also requires ECMP based on label hashing to ignore
> >> reserved labels.
> >
> > You don't want ECMP to hash on PW payload.  With CW and without
> > fat-pw, you don't get ECMP on a PW but you do get ECMP on an LSP
> > containing many PW.
>  
> When the PW payload is IP, the hashing it for ECMP is useful (If the
> payload is not IP then use CW). FAT-PW may not be present in all
> deployments.

I really don't know that any PW are carrying IP.  That was not in the
survey.  L3VPN would be carrying IP.

> >> 2011/3/31 Alexander Vainshtein <[e-mail]>:
> >> > Sriganesh, Curtis and all,
> >> > IMHO and FWIW I think that the only problem with VCCV Type 3 is the
> >> > risk of OAM packets leaking out of T-PE towards the AC (if you
> >> > accidentally set the TTL value too high). To the best of my
> >> > understanding the offset proposal does not solve that - or did I miss
> >> > something?
> >> >
> >> > One way to solve it is to use VCCV Type 1 (clean where applicable).
> >> > Another way would be by using GAL below the PW label (possible but
> >> > messy). If the industry moves towards deprecating Ethernet PWs
> >> > without CW,  it becomes completely unnecessary.
> >> >
> >> > My 2c,
> >> >     Sasha
> >
> > This is the context suggesting a potential problem with VCCV Type 3.
> >
> >> >> -----Original Message-----
> >> >> From: [e-mail] [mailto:[e-mail]] On Behalf Of
> >> >> Sriganesh Kini
> >> >> Sent: Thursday, March 31, 2011 10:09 AM
> >> >> To: [e-mail]
> >> >> Cc: [e-mail]
> >> >> Subject: Re: [PWE3] GAL above and below the PW label
> >> >>
> >> >> All,
> >> >>
> >> >> Just an FYI that a proposal for fate-sharing PW OAM and data given in
> >> >> draft-kini-pwe3-inband-cc-offset
> >> >>
> >> >> Just to summarize the proposal.
> >> >> 1. It does not require GAL in PW. TTL expiry is used to alert the S/T-
> >> >> PE.
> >> >> 2. It is mainly applicable when CW is not used in the PW.
> >> >> 3. It uses a fixed offset (negotiated between PW endpoints) after the
> >> >> label stack before starting the OAM msg.
> >> >> 4. The bytes between the label-stack and the fixed offset is referred
> >> >> to as a pseudoflow header and is filled with byte-values (by the PE)
> >> >> that represent the flow for which OAM is desired. This helps PW OAM
> >> >> and data to fate-share even when the intermediate node looks beyond
> >> >> label stack to do multipath forwarding decisions.
> >> >>
> >> >>
> >> >> On Thu, Mar 31, 2011 at 12:03 AM, Curtis Villamizar <[e-mail]>
> >> >> wrote:
> >> >> >
> >> >> > Loa, Dave, Sasha,
> >> >> >
> >> >> > I've snipped from various posts on the thread that Sasha started.
> >> >>  See
> >> >> > inline.
> >> >> >
> >> >> > In message <[e-mail]>
> >> >> > Loa Andersson writes:
> >> >> >>
> >> >> >> All,
> >> >> >>
> >> >> >> missed a nuance in Sasha subject line.
> >> >> >>
> >> >> >> We have two issues
> >> >> >>
> >> >> >> - where the GAL is placed relative to the PW label, I  believe it is
> >> >> >>    necessary to have the GAL below the PW label.
> >> >> >>
> >> >> >> - whether the the GAL label needs to be bottom of stack or not, it
> >> >> >>    figure that this is really a discussion if it is possible to have
> >> >> >>    a FAT label below the GAL or not. I'm not sure  about my
> >> >> preferences
> >> >> >>    but I think it is possible.
> >> >> >>
> >> >> >> /Loa
> >> >> >
> >> >> > I agree with Loa's summary.  Not putting GAL at the bottom may
> >> >> confuse
> >> >> > some LSR, but putting it above the PW label is likely to be even more
> >> >> > problematic.
> >> >> >
> >> >> >
> >> >> > In message
> >> >> <[e-mail]
> >> >> .se>
> >> >> > David Allan I writes:
> >> >> >>
> >> >> >> It does become a new behavior, GAL with S=3D0. However the
> >> >> combination of G=
> >> >> >> AL being top label at the S-PE and TTL being encoded in the PW label
> >> >> means =
> >> >> >> that fate sharing is broken at every S-PE. Life only gets a little
> >> >> more str=
> >> >> >> ange if there is a FAT label as well...
> >> >> >>
> >> >> >> That being said, GAL as bottom label is broken in any ECMP
> >> >> environment, whi=
> >> >> >> ch is why GAL is a TP construct.
> >> >> >>
> >> >> >> my 2 cents
> >> >> >> D
> >> >> >
> >> >> > Dave,
> >> >> >
> >> >> > If GAL is broken for ECMP, which it is, then all TP OAM is broken.
> >> >> > If all it takes to fix it is simple then lets just fix it.
> >> >> >
> >> >> > This is what we'd have to do.
> >> >> >
> >> >> >  1) Relax the requirement that GAL be at the bottom
> >> >> >
> >> >> >  2) Have the ingress insert GAL in the stack immediately below the
> >> >> >     label for which the measurement is made, keeping the rest of the
> >> >> >     label stack in place.
> >> >> >
> >> >> > The only thing the midpoing LSR at a multipath (LAG, link bundle for
> >> >> > MPLS) has to do is skip over the GAL when hashing, as if the GAL
> >> >> > wasn't there.  That will yield the same hash value and it will
> >> >> > preserve fate-sharing across multipath.
> >> >> >
> >> >> > I prefer that we fix things rather than complain that they are
> >> >> broken.
> >> >> >
> >> >> > Curtis
> >> >> > _______________________________________________
> >> >> > pwe3 mailing list
> >> >> > [e-mail]
> >> >> > https://www.ietf.org/mailman/listinfo/pwe3
> >> >> >
> >> >>
> >> >>
> >> >>
> >> >> --
> >> >> - Sri
> >> >> _______________________________________________
> >> >> pwe3 mailing list
> >> >> [e-mail]
> >> >> https://www.ietf.org/mailman/listinfo/pwe3
> >> > _______________________________________________
> >> > pwe3 mailing list
> >> > [e-mail]
> >> > https://www.ietf.org/mailman/listinfo/pwe3
> >> >
> >>
> >>
> >>
> >> --
> >> - Sri
> >>
> >>
> >
> > _______________________________________________
> > pwe3 mailing list
> > [e-mail]
> > https://www.ietf.org/mailman/listinfo/pwe3