The MPLS-OPS Archive

["Amos Rosenboim" <slick@inter.net.il>]

I personally don't think this is an mpls problem.
I think this is more basic, pri stability issues.
First i would get a stable link between the routers, with the ospf stable
and ipv4 operating , and only then i would go and try to
implement mpls on that link.
As far as stefans remark, i don't know about such a limitation but it's very
much a possibility.
Maybe one of the cisco guys in the list can help in this issue.


Thanks
Amos





----- Original Message -----
From: "alok" <alok.dube@apara.com>
To: "Stefan Oettl" <soettl@freenet.de>; <mpls-ops@mplsrc.com>
Sent: Tuesday, September 10, 2002 3:49 PM
Subject: Re: [MPLS-OPS]: MPLS over ISDN PRI


> i think Amos made a good point on my last mail....its a PE -CE
> connection..so no need to support MPLS over that.
>
> what you cud do is this:
> is ur calling end have a fixed ip address or a pool (shud be)...
>
> assuming u have a seperate ISDN RAS....u wud have some connectivity b/w
the
> PE and the RAS, u cud use acls on the ethernet connecting to the RAS and
put
> it into the right vrf...or mpls VPN...
>
> similarly...if the call terminates on the PE.. and ISDN interfaces cant be
> directly put onto the vrf, u cud use acls to filter traffic based on
source
> ip and put it into the vrf for that mpls vpn...
>
> i know that filters to vrf mapping seems to work on some routers..not sure
> abt cisco though....
>
> -rgds
> A
> ----- Original Message -----
> From: Stefan Oettl <soettl@freenet.de>
> To: Saifulazza <Saifulazza@vads.com>; <mpls-ops@mplsrc.com>
> Sent: Tuesday, September 10, 2002 5:12 PM
> Subject: Re: [MPLS-OPS]: MPLS over ISDN PRI
>
>
> what kind of PE boxes do you use ?
> in case of Cisco's you could configure a backup PE-PE or PE-NAS (NAS will
> need
> MPLS then) GRE tunnel for label switching.
> As far as I know MPLS forwarding on (DDR) ISDN / Dialer interfaces isn't
> supported directly in IOS up to know. Please someone correct me if I'm
wrong
> here ...
> So you could use GRE encapsulation to tunnel the MPLS frames over the
> non-supported circuit.
>
> cheers,
> Stefan
>
> On Tuesday 10 September 2002 11:08, Saifulazza wrote:
> > Hi
> >
> > Has anybody out  have implemented the PE-PE connection  being backup by
> > ISDN PRI ? . I experienced very unstable connection with the OSPF
> adjacency
> > flapping most of the time and the VPN can never be developed .
> >
> > Cheers
> >
> >
> >
> > -----Original Message-----
> > From: fraanro [mailto:fraanro@arrakis.es]
> > Sent: Tuesday, September 10, 2002 4:02 PM
> > To: Enrique Torres
> > Cc: mpls-ops@mplsrc.com
> > Subject: Re: [MPLS-OPS]: tag-switching MTU on c7206 and c3640
> >
> >
> > You should configure the tagswitching MTU in the Fast Ethernet
> > interfaces to be able to hold the MPLS frame with as many shim headers
> > as you need. for example 1508 if you want to labels.
> >
> > mpls mtu 1508
> >
> > under the interface configuration.
> >
> > Rgds.
> > Javier.
> >
> > ----- Mensaje Original -----
> > Remitente: Enrique Torres <entorresa@yahoo.com>
> > Fecha: Lunes, Septiembre 9, 2002 8:28 pm
> > Asunto: [MPLS-OPS]: tag-switching MTU on c7206 and c3640
> >
> > > Hi all
> > >
> > > I started tag-switching between 2 nodes of my network
> > > but there is a problem with the MTU and the packet
> > > size after tag imposition.
> > >
> > > My configuration in each node is it
> > >
> > > I have a Cisco7206 (IOS 12.1(1)E), it is connected
> > > with 6 Cisco 3640 (remote cells) through E1 circuits.
> > > Between the C3640 and C7206 I run tag switching
> > > because I have configured VPN's. In all these routers
> > > is configured "Path MTU Discovery" and our MTU between
> > > the c3640 and c7206 is 1530.
> > >
> > > In this configuration I dont have any problem to send
> > > information between clients connected in our c3640's.
> > >
> > > But when I connect this node with a second node, I
> > > must connect the c7206 through a fast ethernet port
> > > and run tag-switching over these ports. In this
> > > situation I started to have problems with http, ftp or
> > > any kind of traffic with large size of packet (greater
> > > than 1500bytes).
> > >
> > > I believe that I have a problem with the MTU after
> > > second tag imposition (the impositon between c7206's)
> > >
> > > Please, could you help me to solve this.
> > >
> > > Thanks,
> > >       Enrique.
> > >
> > >
> > >
> > > =====
> > > Saludos,
> > >       Enrique
> > >
> > > __________________________________________________
> > > Do You Yahoo!?
> > > Yahoo! Finance - Get real-time stock quotes
> > > http://finance.yahoo.com
> > >
> > > -------
> > > The MPLS-OPS Mailing List
> > > Subscribe/Unsubscribe:  http://www.mplsrc.com/mplsops.shtml
> > > Archive: http://www.mplsrc.com/mpls-ops_archive.shtml
> >
> > -------
> > The MPLS-OPS Mailing List
> > Subscribe/Unsubscribe:  http://www.mplsrc.com/mplsops.shtml
> > Archive: http://www.mplsrc.com/mpls-ops_archive.shtml
> >
> > -------
> > The MPLS-OPS Mailing List
> > Subscribe/Unsubscribe:  http://www.mplsrc.com/mplsops.shtml
> > Archive: http://www.mplsrc.com/mpls-ops_archive.shtml
>
> -------
> The MPLS-OPS Mailing List
> Subscribe/Unsubscribe:  http://www.mplsrc.com/mplsops.shtml
> Archive: http://www.mplsrc.com/mpls-ops_archive.shtml
>
>
>
> -------
> The MPLS-OPS Mailing List
> Subscribe/Unsubscribe:  http://www.mplsrc.com/mplsops.shtml
> Archive: http://www.mplsrc.com/mpls-ops_archive.shtml

-------
The MPLS-OPS Mailing List
Subscribe/Unsubscribe:  http://www.mplsrc.com/mplsops.shtml
Archive: http://www.mplsrc.com/mpls-ops_archive.shtml