For the most part, the redistribution of routes into BGP is a straightforward process. As is the case with other routing protocols, this is performed using the redistribute [protocol]command. By default, BGP will simply use the IGP metric when external routing information is injected into BGP. This negates the need to specify a seed metric or assign a different route metric during redistribution. There are, however, some caveats which you should be familiar with when troubleshooting route advertisement following redistribution.
One common issue is the advertisement of OSPF routes. By default, when OSPF is redistributed into BGP only internal OSPF and external Type 1 routes will be redistributed and advertised. This is often a point of confusion because by default external OSPF routes in Cisco IOS software are external Type 2. You can verify which OSPF route type are being redistributed into BGP by either looking at the device configuration or using the show ip route command, a sample output of which is provided below:
| R4#show ip route 192.168.1.0 255.255.255.0 Routing entry for 192.168.1.0/24 Known via “ospf 4”, distance 110, metric 20, type extern 2, forward metric 1 Redistributing via bgp 4 Last update from 10.0.0.1 on FastEthernet0/0, 00:10:23 ago Routing Descriptor Blocks: * 10.0.0.1, from 1.1.1.1, 00:10:23 ago, via FastEthernet0/0 Route metric is 20, traffic share count is 1 |
NOTE: Although the output above shows that the Type 2 external is being redistributed via BGP, the route is not imported into the BGP RIB because only Type 1 externals will be redistributed by default. This can be validated using the show ip bgp command as follows:
| R4#show ip bgp 192.168.1.0 255.255.255.0 % Network not in table |
When redistributing OSPF routes into BGP, you must specify the OSPF route types that will be redistributed into BGP using the redistribute ospf [process ID] match <external 1|2> <internal> <nssa-external 1|2> router configuration command. Following this, you can again validate the implementation by looking at the router configuration or using the show ip protocols command. For example, if BGP was configured to redistribute all internal and Type 2 externals (Type 5 and Type 7), the show ip route command would show the following:
| R4#show ip route 192.168.1.0 255.255.255.0 Routing entry for 192.168.1.0/24 Known via “ospf 4”, distance 110, metric 20, type extern 2, forward metric 1 Redistributing via bgp 4 Advertised by bgp 4 match internal external 2 nssa-external 2 Last update from 10.0.0.1 on FastEthernet0/0, 00:13:14 ago Routing Descriptor Blocks: * 10.0.0.1, from 1.1.1.1, 00:13:14 ago, via FastEthernet0/0 Route metric is 20, traffic share count is 1 |
Because the route type is included in the redistribution configuration, the route is installed into the BGP RIB and, assuming no filtering configuration, will be advertised to neighbors:
| R4#show ip bgp 192.168.1.0 255.255.255.0 BGP routing table entry for 192.168.1.0/24, version 63 Paths: (1 available, best #1, table Default-IP-Routing-Table) Advertised to update-groups: 1 Local 10.0.0.1 from 0.0.0.0 (4.4.4.4) Origin incomplete, metric 20, localpref 100, weight 32768, valid, sourced, best |
NOTE: In the output above, notice that the BGP next hop is set to 10.0.0.1, which is the OSPF next hop address. In addition, the BGP metric for this prefix is the same as the OSPF metric.
Another common redistribution problem pertains to the redistribution of BGP routes into an IGP, such as EIGRP or OSPF. By default, only external BGP routes are redistributed. This default behavior is used to avoid routing loops within the interior network. When redistributing BGP routes into an IGP, you must use the bgp redistribute-internal command to redistribute iBGP prefixes into the IGP.
NOTE: The redistribution of BGP into IGPs is not recommended. However, if it must be performed, be sure to use route filters to allow only the explicit prefixes that should be redistributed into the IGP to be imported. Do not blindly redistribute BGP into any IGP.