[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: [PWE3] PWE3 question
Sasha,
ER-LSPs work by DoD label assignment. And the ERO identifies the path.
VC FEC is not needed to be processed in the intermediate LSRs at all.
-Shahram
> -----Original Message-----
> From: Sasha Vainshtein [mailto:Sasha@AXERRA.com]
> Sent: Thursday, May 09, 2002 2:58 AM
> To: Shahram Davari; 'Andrew G. Malis'
> Cc: 'pwe3@ietf.org'; Danny McPherson (E-mail)
> Subject: RE: [PWE3] PWE3 question
>
>
> Shahram,
>
> IMHO, if you use only one label for a PW, it is this label that
> is switched through the path from ingress to egress.
> Aand if the routesr do not recognize the VC FEC, how should
> they assign and switch labels associated with this FEC?
>
> With best regards,
> Sasha Vainshtein
> email: sasha@axerra.com <mailto:sasha@axerra.com>
> tel: +972-3-7659993 (office)
> +972-8-9254948 (res.)
> +972-58-674833 (cell.)
>
>
>
> > -----Original Message-----
> > From: Shahram Davari [mailto:Shahram_Davari@pmc-sierra.com]
> > Sent: Wednesday, May 08, 2002 9:05 PM
> > To: 'Sasha Vainshtein'; 'Andrew G. Malis'
> > Cc: 'pwe3@ietf.org'; Danny McPherson (E-mail)
> > Subject: RE: [PWE3] PWE3 question
> >
> >
> > Sasha,
> >
> > I don't think in this case all the routers in the PSN
> > need to recognize the VC FEC. You could set the U/F bits to 1
> > in VC FEC, therefore no need for the interior routers to
> > recognize the VC FEC.
> >
> >
> > -Shahram
> >
> >
> >
> > > -----Original Message-----
> > > From: Sasha Vainshtein [mailto:Sasha@AXERRA.com]
> > > Sent: Wednesday, May 08, 2002 1:59 PM
> > > To: 'Andrew G. Malis'; Shahram Davari
> > > Cc: 'pwe3@ietf.org'; Danny McPherson (E-mail)
> > > Subject: RE: [PWE3] PWE3 question
> > >
> > >
> > > Andy and Shahram,
> > >
> > > IMHO using only one label would, regardless
> > > of the other issues, violate one of the requirements
> > > of the PWE3 Charter:
> > >
> > > <quote>
> > > PWE3 will not exert controls on the underlying PSN,
> > > but only function at the endpoints of the pseudo wire,
> > > though the endpoints may be configured to set diffserv
> > > codepoints in the tunneling datagrams.
> > > <end quote>
> > >
> > > Building an explicit TE tunnel for a single PW means to me
> > > exerting a control over the PSN (e.g., all the routers in the PSN
> > > should recognize VC FECs etc.).
> > >
> > > In other words, it probably can be done, but it would not be PWE3.
> > >
> > > With best regards,
> > > Sasha Vainshtein
> > > email: sasha@axerra.com <mailto:sasha@axerra.com>
> > > tel: +972-3-7659993 (office)
> > > +972-8-9254948 (res.)
> > > +972-58-674833 (cell.)
> > >
> > >
> > >
> > > > -----Original Message-----
> > > > From: Andrew G. Malis [mailto:Andy.Malis@VivaceNetworks.com]
> > > > Sent: Wednesday, May 08, 2002 5:51 PM
> > > > To: Shahram Davari
> > > > Cc: 'Andrew G. Malis'; 'pwe3@ietf.org'
> > > > Subject: RE: [PWE3] PWE3 question
> > > >
> > > >
> > > > Shahram,
> > > >
> > > > Yeah, you would have to treat each VC as a separate TE
> > > tunnel and use
> > > > either RSVP-TE or CR-LDP for signaling (suitably extended
> > > > along Martini
> > > > lines for VC type and ID).
> > > >
> > > > Cheers,
> > > > Andy
> > > >
> > > > -------
> > > >
> > > > At 5/8/2002 08:33 AM -0700, Shahram Davari wrote:
> > > > >Andy,
> > > > >
> > > > >Sure I understand the scaling issue. But ignoring the
> > scalability,
> > > > >I want to know if it can be done and if so how?
> > > > >
> > > > >It seems that you may not be able to use LDP any more,
> > > because VC FEC
> > > > >is meaningless to the interior LSRs. But it seems that you
> > > could use
> > > > >ER-LSPs and therefore do TE. Right?
> > > > >
> > > > >Yours,
> > > > >-Shahram
> > > > >
> > > > >
> > > > > > -----Original Message-----
> > > > > > From: Andrew G. Malis [mailto:Andy.Malis@vivacenetworks.com]
> > > > > > Sent: Tuesday, May 07, 2002 6:12 PM
> > > > > > To: Shahram Davari
> > > > > > Cc: 'pwe3@ietf.org'
> > > > > > Subject: Re: [PWE3] PWE3 question
> > > > > >
> > > > > >
> > > > > > Shahram,
> > > > > >
> > > > > > Well, it wouldn't scale very well since every switch along
> > > > > > the path is
> > > > > > involved in the signaling for each VC, and you can't
> > do traffic
> > > > > > engineering, so I would say that's pretty much a
> non-starter.
> > > > > >
> > > > > > Cheers,
> > > > > > Andy
> > > > > >
> > > > > > At 5/7/2002 12:02 PM -0700, Shahram Davari wrote:
> > > > > > >Hi,
> > > > > > >
> > > > > > >Can we implement PWE3 with only one label (i.e., only
> > > VC label)?
> > > > > > >
> > > > > > >-Shahram
> > > > > > >
> > > > > > >_______________________________________________
> > > > > > >pwe3 mailing list
> > > > > > >pwe3@ietf.org
> > > > > > >https://www1.ietf.org/mailman/listinfo/pwe3
> > > > > >
> > > >
> > > >
> > > > _______________________________________________
> > > > pwe3 mailing list
> > > > pwe3@ietf.org
> > > > https://www1.ietf.org/mailman/listinfo/pwe3
> > > >
> > >
> >
>
_______________________________________________
pwe3 mailing list
pwe3@ietf.org
https://www1.ietf.org/mailman/listinfo/pwe3