Mailing List Archive

[nsp] MPLS: a couple of questions
Hello,
I would like to know if there are limitations for MPLS and OSPF: do
all routers that have to share a vpn need to be on the same OSPF area?


Is it possible to have MPLS over MPLS? In other words, can a customer
of mine that sees his routers connected with VPNs (without seeing my
backbone) in turn sell VPNs to _his_ customers ?



(cust_of_cust)--(my customer)--(my backbone)--(my customer)--(cust_of_cust)
( pop A ) ( pop A ) ( pop B ) ( pop B )

The general idea here is that "my customer" is a separate branch of
our company and we (the backbone) don't want them to mess either with
our OSPF area 0 or with our existing mpls configuration.

Thanks

Pf



--

This E-Mail has NOT been scanned for viruses, so it may be NOT CLEAN.
Wash your hands before meals.

-------------------------------------------------------------------------------
Pierfrancesco Caci | System Administrator @ seabone.net
p.caci@seabone.net | Telecom Italia S.p.A. - International Wholesale Svcs
Linux paperino 2.4.20-pre5 #1 Tue Sep 10 16:45:34 CEST 2002 i686 unknown unknown GNU/Linux
Re: [nsp] MPLS: a couple of questions [ In reply to ]
Pierfrancesco,

At 10:35 AM 10/23/2002 +0200, Pierfrancesco Caci wrote:

>Hello,
>I would like to know if there are limitations for MPLS and OSPF: do
>all routers that have to share a vpn need to be on the same OSPF area?

There is no need for a single area if you use LDP or TDP to signal the LSPs
between the PE routers. If you use MPLS TE instead, you have to make sure
to run a level of code that supports inter-area tunnels if you don't want
the one area restriction to apply.

See the following URL for more information on MPLS-TE inter-area tunnels:

http://www.cisco.com/univercd/cc/td/doc/product/software/ios120/120newft/120limit/120s/120s22/fs_areat.htm


>Is it possible to have MPLS over MPLS? In other words, can a customer
>of mine that sees his routers connected with VPNs (without seeing my
>backbone) in turn sell VPNs to _his_ customers ?

Yes, it is possible. See the following URL to get more information on
Carrier Supporting Carrier:

http://www.cisco.com/univercd/cc/td/doc/product/software/ios120/120newft/120limit/120s/120s22/csc22.htm



>(cust_of_cust)--(my customer)--(my backbone)--(my customer)--(cust_of_cust)
>( pop A ) ( pop A ) ( pop B ) ( pop B )
>
>The general idea here is that "my customer" is a separate branch of
>our company and we (the backbone) don't want them to mess either with
>our OSPF area 0 or with our existing mpls configuration.
>
>Thanks
>
>Pf
>
>
>
>--
>
> This E-Mail has NOT been scanned for viruses, so it may be NOT CLEAN.
> Wash your hands before meals.
>
>-------------------------------------------------------------------------------
> Pierfrancesco Caci | System Administrator @ seabone.net
> p.caci@seabone.net | Telecom Italia S.p.A. - International Wholesale
> Svcs
> Linux paperino 2.4.20-pre5 #1 Tue Sep 10 16:45:34 CEST 2002 i686
> unknown unknown GNU/Linux
>
>_______________________________________________
>cisco-nsp mailing list real_name)s@puck.nether.net
>http://puck.nether.net/mailman/listinfo/cisco-nsp
>archive at http://puck.nether.net/pipermail/cisco-nsp/

Harold Ritter, CCIE 4168
Advanced Network Services - ISP East
Cisco Systems
300 Apollo Drive
Chelmsford, MA 01824 USA
Phone: 978 497 3129
Fax: 978 497 3129
Cisco Systems- "Empowering the Internet Generation."
RE: [nsp] MPLS: a couple of questions [ In reply to ]
>I would like to know if there are limitations for MPLS and OSPF: do
>all routers that have to share a vpn need to be on the same OSPF area?

No, you may be mixing this up with MPLS traffic engineering which does have
more issues crossing OSPF areas.


Matt.

------------------------------------------------------------------------------
Live Life in Broadband
www.telewest.co.uk


The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material.
Statements and opinions expressed in this e-mail may not represent those of the company. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender immediately and delete the material from any computer.


==============================================================================
Re: [nsp] MPLS: a couple of questions [ In reply to ]
> Hello,
> I would like to know if there are limitations for MPLS and OSPF: do
> all routers that have to share a vpn need to be on the same OSPF area?
>

I think you mean if the CEs in one VPN have to be in the same ospf area,

http://www.cisco.com/warp/public/121/mpls_ospf1.html


HTH
--kent
Re: [nsp] MPLS: a couple of questions [ In reply to ]
> Is it possible to have MPLS over MPLS? In other words, can a customer
> of mine that sees his routers connected with VPNs (without seeing my
> backbone) in turn sell VPNs to _his_ customers ?

You can do this with MPLS CsC, have a look at the following URL

http://www.cisco.com/en/US/products/sw/iosswrel/ps1839/products_feature_guid
e09186a0080087cd7.html

Ilker


>
>
>
> (cust_of_cust)--(my customer)--(my backbone)--(my
customer)--(cust_of_cust)
> ( pop A ) ( pop A ) ( pop B ) ( pop
)
>
> The general idea here is that "my customer" is a separate branch of
> our company and we (the backbone) don't want them to mess either with
> our OSPF area 0 or with our existing mpls configuration.
>
> Thanks
>
> Pf
>
>
>
> --
>
> This E-Mail has NOT been scanned for viruses, so it may be NOT CLEAN.
> Wash your hands before meals.
>
> --------------------------------------------------------------------------
-----
> Pierfrancesco Caci | System Administrator @ seabone.net
> p.caci@seabone.net | Telecom Italia S.p.A. - International Wholesale
Svcs
> Linux paperino 2.4.20-pre5 #1 Tue Sep 10 16:45:34 CEST 2002 i686
unknown unknown GNU/Linux
>
> _______________________________________________
> cisco-nsp mailing list real_name)s@puck.nether.net
> http://puck.nether.net/mailman/listinfo/cisco-nsp
> archive at http://puck.nether.net/pipermail/cisco-nsp/
>
RE: [nsp] MPLS: a couple of questions [ In reply to ]
Matt,

At 03:42 PM 10/23/2002 +0100, Matt Ryan wrote:
> >I would like to know if there are limitations for MPLS and OSPF: do
> >all routers that have to share a vpn need to be on the same OSPF area?
>
>No, you may be mixing this up with MPLS traffic engineering which does have
>more issues crossing OSPF areas.

For your information, MPLS TE and MPLS VPN are not mutually exclusive. MPLS
TE can be used to signal the LSP between the PE routers.


>Matt.
>
>------------------------------------------------------------------------------
>Live Life in Broadband
>www.telewest.co.uk
>
>
>The information transmitted is intended only for the person or entity to
>which it is addressed and may contain confidential and/or privileged material.
>Statements and opinions expressed in this e-mail may not represent those
>of the company. Any review, retransmission, dissemination or other use of,
>or taking of any action in reliance upon, this information by persons or
>entities other than the intended recipient is prohibited. If you received
>this in error, please contact the sender immediately and delete the
>material from any computer.
>
>
>==============================================================================
>
>_______________________________________________
>cisco-nsp mailing list real_name)s@puck.nether.net
>http://puck.nether.net/mailman/listinfo/cisco-nsp
>archive at http://puck.nether.net/pipermail/cisco-nsp/

Harold Ritter, CCIE 4168
Advanced Network Services - ISP East
Cisco Systems
300 Apollo Drive
Chelmsford, MA 01824 USA
Phone: 978 497 3129
Fax: 978 497 3129
Cisco Systems- "Empowering the Internet Generation."
RE: [nsp] MPLS: a couple of questions [ In reply to ]
>For your information, MPLS TE and MPLS VPN are not mutually exclusive. MPLS

>TE can be used to signal the LSP between the PE routers.

Indeed, but I have a feeling that the poster was not intending to mesh his
VPN PE routers with TE LSPs.

I'd be interested to hear of a large scale MPLS VPN deployment that does use
TE LSPs between the PE nodes. I'm sure numerous will have some TE paths in
the core with LDP/TDP running over them to provide the end-to-end
signalling, but do networks run TE out to the PE edge?


Matt.

------------------------------------------------------------------------------
Live Life in Broadband
www.telewest.co.uk


The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material.
Statements and opinions expressed in this e-mail may not represent those of the company. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender immediately and delete the material from any computer.


==============================================================================
RE: [nsp] MPLS: a couple of questions [ In reply to ]
Matt,

Commets in-line.

At 10:34 AM 10/24/2002 +0100, Matt Ryan wrote:
> >For your information, MPLS TE and MPLS VPN are not mutually exclusive. MPLS
>
> >TE can be used to signal the LSP between the PE routers.
>
>Indeed, but I have a feeling that the poster was not intending to mesh his
>VPN PE routers with TE LSPs.
>
>I'd be interested to hear of a large scale MPLS VPN deployment that does use
>TE LSPs between the PE nodes. I'm sure numerous will have some TE paths in
>the core with LDP/TDP running over them to provide the end-to-end
>signalling, but do networks run TE out to the PE edge?

I agree that this is not as common as using LDP or TDP. I just wanted to
make sure that people understood that TE and VPN can be used together.


>Matt.
>
>------------------------------------------------------------------------------
>Live Life in Broadband
>www.telewest.co.uk
>
>
>The information transmitted is intended only for the person or entity to
>which it is addressed and may contain confidential and/or privileged material.
>Statements and opinions expressed in this e-mail may not represent those
>of the company. Any review, retransmission, dissemination or other use of,
>or taking of any action in reliance upon, this information by persons or
>entities other than the intended recipient is prohibited. If you received
>this in error, please contact the sender immediately and delete the
>material from any computer.
>
>
>==============================================================================

Harold Ritter, CCIE 4168
Advanced Network Services - ISP East
Cisco Systems
300 Apollo Drive
Chelmsford, MA 01824 USA
Phone: 978 497 3129
Fax: 978 497 3129
Cisco Systems- "Empowering the Internet Generation."
RE: [nsp] MPLS: a couple of questions [ In reply to ]
> >For your information, MPLS TE and MPLS VPN are not mutually
> exclusive. MPLS
>
> >TE can be used to signal the LSP between the PE routers.
>
> Indeed, but I have a feeling that the poster was not
> intending to mesh his
> VPN PE routers with TE LSPs.
>
> I'd be interested to hear of a large scale MPLS VPN
> deployment that does use
> TE LSPs between the PE nodes. I'm sure numerous will have

Well... I'd be interested to hear of any (stable, successful) large
scale deployment that uses LDP... ;-)

Thanks,

Chris

> some TE paths in
> the core with LDP/TDP running over them to provide the end-to-end
> signalling, but do networks run TE out to the PE edge?
>
>
> Matt.
>
> --------------------------------------------------------------
> ----------------
> Live Life in Broadband
> www.telewest.co.uk
>
>
> The information transmitted is intended only for the person
> or entity to which it is addressed and may contain
> confidential and/or privileged material.
> Statements and opinions expressed in this e-mail may not
> represent those of the company. Any review, retransmission,
> dissemination or other use of, or taking of any action in
> reliance upon, this information by persons or entities other
> than the intended recipient is prohibited. If you received
> this in error, please contact the sender immediately and
> delete the material from any computer.
>
>
> ==============================================================
> ================
>
> _______________________________________________
> cisco-nsp mailing list real_name)s@puck.nether.net
> http://puck.nether.net/mailman/listinfo/cisco-nsp
> archive at http://puck.nether.net/pipermail/cisco-nsp/
>
RE: [nsp] MPLS: a couple of questions [ In reply to ]
>Well... I'd be interested to hear of any (stable, successful) large
>scale deployment that uses LDP... ;-)

Depends what you would consider as large I guess. I know of many in the UK
that are live (ourselves included) but then the UK is only as big as some US
states...


Matt.

------------------------------------------------------------------------------
Live Life in Broadband
www.telewest.co.uk


The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material.
Statements and opinions expressed in this e-mail may not represent those of the company. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender immediately and delete the material from any computer.


==============================================================================
RE: [nsp] MPLS: a couple of questions [ In reply to ]
We are using LDP across the COLT AS8220 backbone. Its been stable until
we tried to upgrade to: 12.0-21.ST4 which is probably the most broken
version of IOS that I've ever had to deal with, 12.0-17.ST5 is pretty
good but doesn't have the latest DDOS tracking commands.
Just be wary of traceroute when you make the change, it doesn't work
quite
how it should and some users get spooked when the first hop reports the
total RTT to
the endpoint rather than the first hop. [.Personally I think this is a
good
thing as traceroute output is one of the most wrongly interpreted
network
tools ever].

Regards,
Neil.

> -----Original Message-----
> From: cisco-nsp-admin@puck.nether.net
> [mailto:cisco-nsp-admin@puck.nether.net] On Behalf Of Matt Ryan
> Sent: 25 October 2002 18:48
> To: 'Chris Whyte'
> Cc: 'cisco-nsp@puck.nether.net'
> Subject: RE: [nsp] MPLS: a couple of questions
>
>
> >Well... I'd be interested to hear of any (stable, successful) large
> >scale deployment that uses LDP... ;-)
>
> Depends what you would consider as large I guess. I know of
> many in the UK that are live (ourselves included) but then
> the UK is only as big as some US states...
>
RE: [nsp] MPLS: a couple of questions [ In reply to ]
On Mon, 28 Oct 2002, Neil J. McRae wrote:

> We are using LDP across the COLT AS8220 backbone. Its been stable until
> we tried to upgrade to: 12.0-21.ST4 which is probably the most broken
> version of IOS that I've ever had to deal with, 12.0-17.ST5 is pretty
> good but doesn't have the latest DDOS tracking commands.

We've achieved good stability with 12.0(21)ST2 (GSRs, ISIS, BGP etc...).
The only issue we've had is ttl not being decremented before a packet
is tagged. The TAC thinks that CSCdx14037 is causing this but I don't
know...

What kind of issues did you run into with 12.0(21)ST4?

--
Visit: http://www.stenling.nu/contact.html for contact info.
EOF
Re: [nsp] MPLS: a couple of questions [ In reply to ]
12.0(21)ST4 is the first release under which was got the new
OC-192 E4+ cards to support MPLS with LDP. Under 12.0(22)S1 and other
releases we had forwarding problems whenever MPLS was enabled. Traffic
engineering with fast re-route is working properly on a combination of E2,
E3, and E4+ cards.


On Tue, Oct 29, 2002 at 09:51:39AM +0100, Jonas Stenling wrote:
> On Mon, 28 Oct 2002, Neil J. McRae wrote:
>
> > We are using LDP across the COLT AS8220 backbone. Its been stable until
> > we tried to upgrade to: 12.0-21.ST4 which is probably the most broken
> > version of IOS that I've ever had to deal with, 12.0-17.ST5 is pretty
> > good but doesn't have the latest DDOS tracking commands.
>
> _______________________________________________
> cisco-nsp mailing list real_name)s@puck.nether.net
> http://puck.nether.net/mailman/listinfo/cisco-nsp
> archive at http://puck.nether.net/pipermail/cisco-nsp/

--
=========================================================================
Clinton Work clinton@scripty.com
Calgary, Alberta