Route reflection and MPLS VPN 14.4

As I was going through my verification process I noticed that R4 would not install the vpnv4 pfxs unless R6 was configured as a route reflector client, the same goes for R5. In a classical BGP setup R4 would install the routes but would not pass it onto its peers. I did complete the task but I haven't been able to move on because I've been researching the "why". 

Does anyone know the "why"?

 

Here is the result of having the route reflector client configured for both R5 & 6...

 


Rack1R4#sh bgp vpnv4 unicast all 

BGP table version is 55, local router ID is 150.1.4.4

Status codes: s suppressed, d damped, h history, * valid, > best, i - internal,

              r RIB-failure, S Stale

Origin codes: i - IGP, e - EGP, ? - incomplete

 

   Network          Next Hop            Metric LocPrf Weight Path

Route Distinguisher: 100:1

*>i155.1.58.0/24    150.1.5.5                0    100      0 ?

*>i155.1.67.0/24    150.1.6.6                0    100      0 ?

*>i192.168.7.0      150.1.6.6                0    100      0 ?

Route Distinguisher: 100:2

*>i155.1.5.0/24     150.1.5.5                0    100      0 ?

*>i155.1.76.0/24    150.1.6.6                0    100      0 ?

*>i172.16.7.0/24    150.1.6.6                0    100      0 ?

 

 

Here it is without route reflector client configured...

 


Rack1R4#sh bgp vpnv4 unicast all 

 

Rack1R4#

 

 

 

Comments

  • Edwinrg00,

    If route reflection is configured under the default (ipv4) address family, that only applies to the ipv4 address family - you probably already know that.

    There is by default a "bgp default route-target filter" under the bgp process that will discard prefixes with unknown route-targets. Configuring peers as route-reflector-clients or applying "no bgp default route-target filter" under the bgp process will permit prefixes with unknown route-targets (not configured in a local vrf) to be passed on.

    The same iBGP peering rules apply under the vpnv4 address family - prefixes learned from iBGP peers will not be passed on to other iBGP peers unless route reflection is configured.

  • Darrell,

    I understand the "BGP default route-target filter" but I thought this only applied to the PE routers. My understanding is that the RR will store all bgp routes because it doesn't know which RTs the PEs permit or deny.

    Sent from my HTC smartphone


    From: DarrellEscola <[email protected]>
    Sent: Sunday, March 21, 2010 10:31 PM
    To: [email protected]
    Subject: Re: [iewb-rs-vol1-v5-mpls] Route reflection and MPLS VPN 14.4

    Edwinrg00,

    If route reflection is configured under the default (ipv4) address family, that only applies to the ipv4 address family - you probably already know that.

    There is by default a "bgp default route-target filter" under the bgp process that will discard prefixes with unknown route-targets. Configuring peers as route-reflector-clients or applying "no bgp default route-target filter" under the bgp process will permit prefixes with unknown route-targets (not configured in a local vrf) to be passed on.

    The same iBGP peering rules apply under the vpnv4 address family - prefixes learned from iBGP peers will not be passed on to other iBGP peers unless route reflection is configured.




    Internetwork Expert - The Industry Leader in CCIE Preparation
    http://www.internetworkexpert.com

    Subscription information may be found at:
    http://www.ieoc.com/forums/ForumSubscriptions.aspx
  • Edwinrg00,

    You are correct - route reflection will override the default route-target filter for configured clients, but the "no bgp default route-target filter" command is needed on Autonomous System Border Routers:

    Usage Guidelines

    Use the bgp default route-target filter command to control the distribution of Virtual Private Network (VPN) routing information through the list of VPN route-target communities.

    When you use the no form of this command, all received VPN-IPv4 routes are accepted by the configured router. Accepting VPN-IPv4 routes is the desired behavior for a router configured as an autonomous system border edge router or as a customer edge (CE) BGP border edge router.

    If you configure the router for BGP route-target community filtering, all received exterior BGP (EBGP) VPN-IPv4 routes are discarded when those routes do not contain a route-target community value that matches the import list of any configured VPN routing/forwarding instances (VRFs). This is the desired behavior for a router configured as a provider edge (PE) router.

    Note: imageThis command is automatically disabled if a PE router is configured as a client of a common VPN-IPv4 route reflector in the autonomous system.

    (From the 12.3 Command Reference)

    Darrell,

    I understand the "BGP default route-target filter" but I thought this only applied to the PE routers. My understanding is that the RR will store all bgp routes because it doesn't know which RTs the PEs permit or deny.

    Sent from my HTC smartphone

  • Darrell,

    I just had the opportunity to test the "No bgp default route-target filter" and your response did clear it up for me.

    I guess I didn't stop to realize that R4, besides being the RR is also setup to be a PE router. Not having R5 & R6 configured as RR clients R4 doesn't install the vpnv4 routes by default because the RT were not defined. By configuring  "No bgp default route-target filter" overrides this default and installs the vpnv4 routes.

     

    Thanks

  • Edwinrg00,

    We can all learn things here - sometimes by reading other posts, and sometimes by researching so we can come up with an answer...

Sign In or Register to comment.