[c-nsp] Dual-Homed VPLS

Tim Durack tdurack at gmail.com
Fri Jul 27 09:08:16 EDT 2007


Disclaimer: I've only read about this.

If you can do H-VPLS with u-PE/n-PE Cisco talks about "EE H-VPLS
Pseudo-n-PE Redundancy" being a way to avoid some of the
loop-avaoidance issues.

See this link:

http://cisco.com/en/US/products/hw/routers/ps368/products_white_paper09186a00801f6084.shtml

Looks like an interesting idea to me. Anyone actually doing this?

Tim:>

On 7/27/07, Peter Krupl <peter.krupl at ventelo.dk> wrote:
> Hi,
>
> I have looked at this issue too, but your solution has one major flaw...
>
> Q:
> What would happen if the VPLS circuits go down in the core network,
> And then came back up ?
> A: You have a loop, until spanning tree notices....
>
> Flexlink seems more usable....
>
>
> Med venlig hilsen/Kind regards
> Peter Åris Krüpl
> Netværksspecialist
>
> -----Oprindelig meddelelse-----
> Fra: cisco-nsp-bounces at puck.nether.net [mailto:cisco-nsp-bounces at puck.nether.net] På vegne af alaerte.vidali at nsn.com
> Sendt: 25. juli 2007 17:33
> Til: helmwork at ruraltel.net
> Cc: cisco-nsp at puck.nether.net
> Emne: Re: [c-nsp] Dual-Homed VPLS
>
> Hi Eric,
>
> Exactly.
>
> As PE would forward Spanning tree BPDUs transparently, I am considering
> STP is also an option to block a link.
> For example:
>
>
> CE1-A_(fa-1/1)---------------  PE1=======PE2---------CE2-A (STP ROOT)
>    |                           \         /
>    (fa-1/2)                     \       /
>    |_________________________    \     /
>                                   PE3
>
>
> Considering CE2-A is STP root, CE1-A would receive BPDU from both
> interfaces, would choose fa1/1 as RP and would block fa1/2.
> What do you think?
>
>
> I hope we receive more feedback.
>
>
> Best Regards,
> Alaerte
>
>
>
> -----Original Message-----
> From: ext Eric Helm [mailto:helmwork at ruraltel.net]
> Sent: Wednesday, July 25, 2007 12:15 PM
> To: Vidali Alaerte (NSN - BR/Rio de Janeiro)
> Subject: Re: [c-nsp] Dual-Homed VPLS
>
> Alaerte...
> Are you talking about dual-homed VPLS endpoints? If so, I'd be curious
> to hear what suggestions you receive for this topic. When I looked into
> doing this, it seemed that using Flex-Links was the only viable
> solution.
>
> Regards,
>
> /Eric
>
> alaerte.vidali at nsn.com wrote:
> >  Hi,
> >
> > Do you indicate any reference for this topic?
> >
> > I tried some books like "MPLS Configuration on Cisco IOS Software"
> > (pretty good book) by Lancy and Umesh, but it only touch the subject.
> >
> > Tks,
> > Alaerte
> > _______________________________________________
> > cisco-nsp mailing list  cisco-nsp at puck.nether.net
> > https://puck.nether.net/mailman/listinfo/cisco-nsp
> > archive at http://puck.nether.net/pipermail/cisco-nsp/
> >
> _______________________________________________
> cisco-nsp mailing list  cisco-nsp at puck.nether.net
> https://puck.nether.net/mailman/listinfo/cisco-nsp
> archive at http://puck.nether.net/pipermail/cisco-nsp/
> _______________________________________________
> cisco-nsp mailing list  cisco-nsp at puck.nether.net
> https://puck.nether.net/mailman/listinfo/cisco-nsp
> archive at http://puck.nether.net/pipermail/cisco-nsp/
>


More information about the cisco-nsp mailing list