Comware Based
cancel
Showing results for 
Search instead for 
Did you mean: 

Blackhole Route in FIB

 
Highlighted
Valued Contributor

Blackhole Route in FIB

I'm having reachability issues in an BGP-L3VPN MPLS test environment, and I've noticed that the unreachable prefixes exist in the FIB with the Blackhole flag.

 

[EDITED]

[LAB-PE3] dis ip routing-table vpn-instance TEST-VRF

Destinations : 24 Routes : 29

Destination/Mask Proto Pre Cost NextHop Interface
10.64.8.0/24 BGP 255 0 10.64.2.1 HGE1/1/25
10.64.9.0/24 BGP 255 0 10.64.2.1 HGE1/1/25
10.66.8.0/24 BGP 255 0 10.64.0.8 HGE1/1/25
10.66.9.0/24 BGP 255 0 10.64.0.8 HGE1/1/25

[LAB-PE3] dis fib vpn-instance TEST-VRF

Destination count: 24 FIB entry count: 36

Flag:
U:Useable G:Gateway H:Host B:Blackhole D:Dynamic S:Static
R:Relay F:FRR

Destination/Mask Nexthop Flag OutInterface/Token Label

10.64.8.0/24 10.64.2.68 UBGR HGE1/1/25 24256 
10.64.9.0/24 10.64.2.68 UBGR HGE1/1/25 24256 
10.66.8.0/24 10.64.0.8 UGR 414 24117 
10.66.9.0/24 10.64.0.8 UGR 414 24117

 

  • The above is taken from PE3.
  • PE1, PE2 and PE3 all connect directly to RR1.
  • PE1 is injecting the 10.66 prefixes and peers with the RR using a loopback IP (10.64.0.8).
  • PE2 is injecting the 10.64 prefixes and peers with the RR using the physical interface IP (10.64.2.1).
  • The 10.66 prefixes are reachable, the 10.64 ones are not.
  • Everyone blackholes PE2's routes.
  • PE3 has a FIB next-hop of 10.64.2.68 for PE2s routes which is (PE3's next-hop on the RR)
  • If I change PE2 to "connect-interaface loopback X", then all the problems are resolved.

I can't work out why this is happening.

Any ideas?