Re: OT: GS Archives Search

John,

That makes sense and I think you exactly reached the point in your e-mail
that original designers of the technology did. Let me try to clarify that.

"Aha, if we want 1:1 communication, then RDs are enough. For everything
else, we need RTs. But, that's inconsistent, why not always use RTs and
have a consistent approach, no matter what the topology is?".

So, with that line of thinking, that I entirely made up, but I'm sure was
part of the thought process, you cannot have L3VPN without route-targets.
In Cisco IOS world, you *can* configure VPN without RT and routes will be
propagated to other routers, which will in turn happily filter them out by
default (bgp default route-target filter). Some other vendors, like for
example Juniper, don't even allow you to create a VPN routing instance (aka
VRF) without specifying the route targets.

Going back to your e-mail - yes, absolutely. RD and T solve different
problems, but in a working L3VPN both must be present.

I'm surprised noone brought up another parameter, which could have been
used instead of RT (but isn't) to associate routes with VPNs - an MPLS
label :-).

--
Marko Milivojevic - CCIE #18427 (SP R&S)
Senior CCIE Instructor - IPexpert
On Tue, Mar 27, 2012 at 10:04, John Neiberger <jneiberger_at_gmail.com> wrote:
> I know virtually nothing about RDs and RTs, but when I'm teaching
> stuff that I am familiar with, I like to start out with the problem.
> My mentor Howard Berkowitz is always fond of asking, "What problem are
> you trying to solve?" Everything in networking is a solution to some
> problem; understand the problem and you understand the solution.
>
> If I'm understanding this correctly, RDs might be all that's necessary
> for a basic VPN where customer A never needs to reach Customer B and
> vice versa. RDs solve the problem of needing to create unique routes
> in BGP for VPNs. But if your requirements change, you might run into
> new problems. What if your basic VPNs are working but you suddenly
> develop the need for some customers to see certain routes from other
> customers? As I understand it, that's where Route Targets come in.
> They solve a new and slightly different problem, right?
>
> John
>
> On Tue, Mar 27, 2012 at 10:03 AM, Marko Milivojevic <markom_at_ipexpert.com>
> wrote:
> > I simply cannot add anything to the fantastically simple explanation
> Carlos provided. If you haven't, read his response. Then go back and read
> the one I sent last night.
> >
> > Always question WHY and you will learn. I like your learning approach -
> it's the right way to do it.
> >
> > --
> > Marko Milivojevic - CCIE #18427
> >
> > :: This message was sent from a mobile device. I apologize for errors
> and brevity. ::
> >
> > On Mar 27, 2012, at 5:40, Olayemi Salau <salauolayemi_at_googlemail.com>
> wrote:
> >
> >> Marc...thanks, this is the search page I was looking for.
> >>
> >> I wanted to search the archives to see if someone has asked why MP-BGP
> couldn't stick with RTs to "distinguish" customer routes, by extending
> normal 4 octect addresses with RT and not RDs, this will still achieve a 96
> bit addr that can be used for MPBGP. I understand why RD cant do the job of
> RTs(flexibilty, complexities and load sharing traffic MPLS VPN traffic
> etc.), but I don't get why RTs can't do the job of RDs. I think the simple
> answer to this lies in the design/conceptual decisions. Perhaps RD was
> designed before RTs came into being. Nothing in the RFC 4364 to say this as
> I read.
> >>
> >> Like everyone said, we all know what they are n do. There are some
> design decisions fundamentals that cant be questioned I suppose, or maybe
> they can....like why Area 0 (not Area 1) for OSPF BB and Level2 (not
> Level0) "contiguousness" for ISIS, why does higher priorities in OSPF and
> lower BID in SPT win elections .....etc
> >>
> >> I thank everyone for their explanations of RT/RD. Very much appreciated.
> >>
> >> Yemi
> >>
> >> On Mar 27, 2012, at 6:19, marc edwards <renorider_at_gmail.com> wrote:
> >>
> >>> Answer to your original question:
> >>>
> >>> http://groupstudy.com/cgi-bin/search
> >>>
> >>> Now a great new thread in the archives from our experts as well :)
> >>>
> >>> HTH
> >>>
> >>> Marc
> >>>
> >>> On Mon, Mar 26, 2012 at 10:18 AM, Yemi Salau <salauolayemi_at_yahoo.co.uk
> >wrote:
> >>>
> >>>> Guys,
> >>>>
> >>>> I remember a time where I was able to search the GS archives for
> stuffs.
> >>>> Is this still available today? I want to search out some stuffs on RD
> vs RT.
> >>>>
> >>>> Yemi
> >>>>
> >>>>
> >>>> Blogs and organic groups at http://www.ccie.net
> >>>>
> >>>>
> _______________________________________________________________________
> >>>> Subscription information may be found at:
> >>>> http://www.groupstudy.com/list/CCIELab.html
> >>>
> >>>
> >>> Blogs and organic groups at http://www.ccie.net
> >>>
> >>> _______________________________________________________________________
> >>> Subscription information may be found at:
> >>> http://www.groupstudy.com/list/CCIELab.html
> >>
> >>
> >> Blogs and organic groups at http://www.ccie.net
> >>
> >> _______________________________________________________________________
> >> Subscription information may be found at:
> >> http://www.groupstudy.com/list/CCIELab.html
> >
> >
> > Blogs and organic groups at http://www.ccie.net
> >
> > _______________________________________________________________________
> > Subscription information may be found at:
> > http://www.groupstudy.com/list/CCIELab.html
Blogs and organic groups at http://www.ccie.net