[c-nsp] VRF Route-leaking question
Mihai Tanasescu
mihai at duras.ro
Sun Jan 13 13:13:14 EST 2008
>> The problem:
>
> The configuration below was not copy+pasted from an IOS configuration.
> It's usually a very good idea to do just that, and not try writing it in
> hand. The initial command for configuring a VRF is "ip vrf <name>", not
> just "vrf <name>" and IOS doesn't use shorthand itself.
Sorry for that..I wanted to avoid giving private information from my config.
My scenario only uses local VRFs (VRF-lite) so I thought there was no
need to import from myself.
The idea was to have:
Router Distribution (Client group 1 + exchange routes + default route 1)
---> BW limitting machine 1 layer 2 ---> Router Core
Router Distribution (Client group 2 + exchange routes + default route 2)
---> BW limitting machine 2 layer 2 ---> Router Core
> Try looking at the routing table after leaking, with a "show ip route
> vrf bbb". It should show your Gi1/1 as connected, learned via BGP. Same
> goes for VRF aaa and Gi1/2. You should be able to ping local interfaces
> too.
>
> This difference between a local and non-local address could maybe point
> at something like FIB-problems. What does "show mls cef vrf bbb
> 192.168.1.1" show?
>
The config + info you required:
interface GigabitEthernet1/34
ip vrf forwarding vrf_metro
ip address 86.104.125.9 255.255.255.0
interface Port-channel2
ip vrf forwarding vrf_test
ip address 79.134.32.181 255.255.255.252
ip vrf vrf_test
rd 43930:35137
route-target export 43930:35137 - client1 group
route-target import 43930:43930 - metro
route-target import 43930:65000 - default route 1
ip vrf vrf_metro
rd 43930:43930
route-target export 43930:43930 - metro
route-target import 43930:35137 - client 1 group
route-target import 43930:65400 - client 2 group
The other RT imports are either the VRF with default_route1,
default_route2 or other test scenarios.
Similar imports are configured in the other VRFs.
#sh ip ro vrf vrf_test 86.104.125.0
Routing Table: vrf_test
Routing entry for 86.104.125.0/24
Known via "bgp 43930", distance 20, metric 0 (connected, via
interface), type external
Routing Descriptor Blocks:
* directly connected, via GigabitEthernet1/34
Route metric is 0, traffic share count is 1
#sh ip ro vrf vrf_metro 86.104.125.0
Routing Table: vrf_metro
Routing entry for 86.104.125.0/24
Known via "connected", distance 0, metric 0 (connected, via interface)
Redistributing via bgp 43930
Advertised by bgp 43930
Routing Descriptor Blocks:
* directly connected, via GigabitEthernet1/34
Route metric is 0, traffic share count is 1
#ping vrf vrf_metro 86.104.125.9
Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 86.104.125.9, timeout is 2 seconds:
!!!!!
Success rate is 100 percent (5/5), round-trip min/avg/max = 1/2/4 ms
#ping vrf vrf_metro 86.104.125.2
Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 86.104.125.2, timeout is 2 seconds:
!!!!!
Success rate is 100 percent (5/5), round-trip min/avg/max = 1/1/4 ms
#ping vrf vrf_test 86.104.125.2
Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 86.104.125.2, timeout is 2 seconds:
!!!!!
Success rate is 100 percent (5/5), round-trip min/avg/max = 48/50/52 ms
#ping vrf vrf_test 86.104.125.9
Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 86.104.125.9, timeout is 2 seconds:
..
Success rate is 0 percent (0/2)
and the loop appears in the logs.
The CEF part:
86.104.125.9, 2 etc are from vrf_metro
#show mls cef vrf vrf_test 86.104.125.2
Codes: decap - Decapsulation, + - Push Label
Index Prefix Adjacency
1006 86.104.125.2/32 Gi1/34 , 000e.0cba.8cba
#show mls cef vrf vrf_metro 86.104.125.2
Codes: decap - Decapsulation, + - Push Label
Index Prefix Adjacency
434 86.104.125.2/32 Gi1/34 , 000e.0cba.8cba
and for the IP on the interface: 86.104.125.9
#show mls cef vrf vrf_test 86.104.125.9
Codes: decap - Decapsulation, + - Push Label
Index Prefix Adjacency
444 86.104.125.9/32 receive
#show mls cef vrf vrf_metro 86.104.125.9
Codes: decap - Decapsulation, + - Push Label
Index Prefix Adjacency
432 86.104.125.9/32 receive
Sorry for the long email.
Any idea what's happening or what am I doing wrong ?
-
Mihai
More information about the cisco-nsp
mailing list