When troubleshooting IPv6 route redistribution you should follow the same processes used for IPv4 route redistribution. These include verifying device configurations and checking appropriate filters used during redistribution, if applicable. In addition to these rules, it is also important to remember default routing protocol operation. For example, neither RIPng nor EIGRPv6 will redistribute routes from OSPFv3 if the seed metric is not specified or if the metric is not included in the redistribution configuration. Therefore, instead of focusing on exclusive scenarios pertaining to IPv6 redistribution, remember the following points when both implementing and troubleshooting IPv6 route redistribution:
- With RIPng, if multiple instances are configured on the same router, by default, these instances will share routing information with each other, i.e. with no explicit redistribution configuration, if they use the same default Multicast group and UDP port. This default behavior can be changed by specifying different UDP port numbers for RIPng instances using the port [number] multicast-group [address] router configuration command when configuring the RIPng instances
- By default, when redistributing between dynamic routing protocols, connected routes are not automatically included in redistribution for IPv6. Instead, the include-connected router configuration command must be appended to the end of the redistribute [protocol]router configuration command
- When redistributing between IPv6 routing protocols, any route filtering implemented during redistribution, i.e. distribute lists and route maps, must reference either IPv6 ACLs or IPv6 prefix lists
- When redistributing IPv6 prefixes into OSPFv3, the subnets keyword is not required as IPv6 does not use Classful networks as IPv4
- By default, all local routes, i.e. routes marked with an ‘L’ in the RIB, are not included in the route redistribution for any IPv6 routing protocol
- Only global IPv6 addresses are redistributed. Link-Local addresses, which are used as the next hop address for IPv6 IGPs, cannot redistributed