The MPLS-OPS Archive

Cell Relay Retreat>MPLS-OPS Archive>month:2002-Jul> msg00089



[Date Prev][Date Next][Thread Prev][Thread Next]  
  [Date Index][Thread Index][Author Index][Subject Index]

Re: Cisco Multi-VRF CE or VRF Light

  • From: Rajiv Asati <rajiva@cisco.com>
  • Date: Sat, 27 Jul 2002 18:25:06 -0400
  • Cc: mpls-ops@mplsrc.com
  • Resent-Date: Sat, 27 Jul 2002 19:23:10 -0400
  • To: "M. ELK" <elkou141061@hotmail.com>
  • X-Sender: rajiva@dingdong.cisco.com

Hi,

Please see inline...

At 01:54 PM 7/27/2002, M. ELK wrote:
Hi
 
Ref :
http://www.cisco.com/warp/public/cc/pd/rt/2600/prodlit/1575_pp.htm
 
http://www.cisco.com/univercd/cc/td/doc/product/rtrmgmt/vpnsc/mpls/2_2/prov_gd/
 
 
AA- The first ref does not address clearly how to connect a Multi-VRF CE
to a PE using eBGP .

Well, configuring BGP in the Multi-VRF CE is nearly the same as configuring it on the PE. :)

The 2nd Ref (page 11-9) provide a template .

Yes.

Any doc which list a complete config for Multi-VRF CE using
ebgp to connect to a PE router ??

I am not sure if I understand you correctly.
The page11-9 indeed provides the necessary config needed to configure  the CE router.
Are you asking about the both PE and CE config when eBGP is used as a PE-CE protocol.

 
 
BB-from ref 2 : say for 5 VRF on the CE , is it mean a ebgp session per
VRF between the CE and the PE ??

Yes. So we will end up configuring "address-family ipv4 vrf <name>" for each VRF in the BGP.
So the BGP config on the CE will look like as follows:
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
bgp <AS#1>
!
 address-family ipv4 vrf v1
 neighbor x1 remote-as <AS#2>
 neighbor x1 activate
 !
 address-family ipv4 vrf v2
 neighbor x2 remote-as <AS#2>
 neighbor x2 activate
 !
 address-family ipv4 vrf v3
 neighbor x3 remote-as <AS#2>
 neighbor x3 activate
 !
~~~~~~~~~~~~~~~~~~~~~~~
And the PE's config will look like this:

bgp <AS#2>
!
 address-family ipv4 vrf v1
 neighbor x1 remote-as <AS#1>
 neighbor x1 activate
 !
 address-family ipv4 vrf v2
 neighbor x2 remote-as <AS#1>
 neighbor x2 activate
 !
 address-family ipv4 vrf v3
 neighbor x3 remote-as <AS#1>
 neighbor x3 activate
 !

 
Could be easier if the CE  and PE run a modified eBGP VPNV4 .

Slight correction: PE-CE are not required to run eBGP VPNv4.. rather they run the normal eBGP (ipv4) session. :)
As we know, eBGP VPNv4 could be interpreted as MP-eBGP which might be run in the Inter-AS case.

 a modified in the sense :
At the PE :
Say the 5 VRF are VRf1 ...VRF5
 
 interface s1
 ip Vrf Forwarding Vrf1,Vrf2,Vrf3,Vrf4,Vrf5
 
address-family VPNV4 VRF vrf1,Vrf2,Vrf3,Vrf4, Vrf5
Nei CE remote-as XXX
...
...
 
a VPNV4 context is configured per VRF(s) and an interface could be
linked to several VRF's .
 
For outgoing eBGP update : advertise route from those VRF routing table , no RT ext-community will be exported .
 
For incoming eBGP update : import the route to the correspending
VRF based on the RD value .  
(Those recieved eBGP VPNV4 update are within the context of  the sepcified VRF (Vrf1 ...VRf5) which mean that they are not propagated to
   other PE's ) .

Well, what if the CE is connected to more than one PE in that VRF. :)

 
at the CE : 
 
Addrres-family VPNV4 vrf vrf1,vrf2,vrf3,vrf4,vrf5
nei PE remote-as YYY
....
 
 .
 
In such case their is  no need to have a sub-interface for each VRF between the CE and the PE , Separation of VRF traffic is now based on the label .

Well, I don't think that the above approach is viable in the current implementation, so I will leave it for others to comment. But if the number of subinterfaces is the concern, then GRE tunnel per VRF could be an alternative.

 
(May be the above is something like Inter-AS RFC 2547 option "a" but
 now is VRF's-to-VRF's  instead of VRF-to-VRF still it is not full fledge
as option "c" or option "b" ) 
 
CC- For a Multi-VRF over a CE where it is connected to PE using
  eBGP (one eBGP session per VRF) ,it is not clear the benefit
  (in term of mem and CPU ) of such setup versus the CE runing full PE-functionality . Pls clarify .

Well, multi-VRF is for the CE that is not required to run MPLS.. and only needs basic VRF configuration. Nothing fancy. So VRF concept gets used to provide the isolation of routing information at the CE itself, without involving MPLS.
We could certainly extend the PE functionality at the CE, but then you end up with CsC architecture and then we don't need multiple interfaces on the PE-CE. It really comes down to the requirement and type of box.

Please let me know if I could be of further help.

Cheers,
Rajiv

 
Brgds


Join the world’s largest e-mail service with MSN Hotmail. Click Here
------- The MPLS-OPS Mailing List Subscribe/Unsubscribe: http://www.mplsrc.com/mplsops.shtml Archive: http://www.mplsrc.com/mpls-ops_archive.shtml