[c-nsp] OSPF support to MPLS

Murilo Antonio Pugliese mpugliese at diveo.net.br
Fri Mar 14 08:34:15 EDT 2008


Oliver.

Thanks a lot for you replies. I as aware about MPLS TE restriction with
a multi-area ospf network, ...

Currently, OSPF supports MPLS traffic engineering in only a single area
- typically, the backbone or area 0.
http://books.google.com.br/books?id=-NlJnfxBXqwC&pg=PA265&lpg=PA265&dq=%
22OSPF+supports+MPLS%22&source=web&ots=lNsJdY-f5K&sig=lGoZmPj41jxtRvILeE
VmSZh1FaM&hl=pt-BR

... but right know the environment that I'm considering doesn't consider
employing MPLS TE.

Thanks a lot buddy.

Yours Truly.

MP.


-----Original Message-----
From: Oliver Boehmer (oboehmer) [mailto:oboehmer at cisco.com] 
Sent: Friday, March 14, 2008 9:32 AM
To: Murilo Antonio Pugliese; cisco-nsp at puck.nether.net
Subject: RE: [c-nsp] OSPF support to MPLS

Oliver Boehmer (oboehmer) <> wrote on Friday, March 14, 2008 1:21 PM:

> Murilo Antonio Pugliese <> wrote on Friday, March 14, 2008 1:10 PM:
> 
>> Folks.
>> 
>> I wonder if a MPLS domain may get extended though many OSPF areas 
>> (belonging to the same OSPF process)?
> 
> sure, just make sure your edge devices know the loopback/BGP next-hop 
> for all other PE devices in other areas (i.e. don't summarize the 
> loopback address range at the ABR or don't just advertise a default).
> 

forgot to add one thing: MPLS traffic-engineering in a multi-area
network has several restrictions (as the headends no longer have full
visibility), but if you're just using L2VPN or L3VPN, this will continue
to work..

	oli


More information about the cisco-nsp mailing list