Hi Makar,
Vijay is correct. The AD is a tiebreaker, i.e. a means to resolve a conflict when it happens. With particular respect to AD, the conflict means that two different sources of routing information try to install the same prefix into the routing table. The same prefix means that both the IP address of the network and the subnet mask are identical. Only if this situation arises, the AD decides which routing source (OSPF, EIGRP, static, ...) is more trustworthy and wins. Note also that AD is about installing routes into the routing table, not about using existing routes in the routing table. Once a route is installed in the routing table, using it for routing packets has other criteria that are not influenced by AD.
In your example, no such conflict exists. You have a summary route with the AD=5 and another more specific route (obviously summarized by the previous summary route) with a higher AD. From the perspective of installing both these routes into your routing table, there is no conflict: the summarized route has a different subnet mask (and possibly the network address itself). Such routing entries may be installed into the routing table immediately.
When routing packets, the usual longest prefix match rule applies: you will always use the most specific route in your routing table to route packets. Even if there are overlapping records exactly like the specific subnets and their summary, from among these entries, the most specific route for the destination will be used. There is no concept of AD involved here.
You have asked when can such a scenario happen. Quite easily. Imagine three routers in a row, R1, R2, and R3. Routers R1 and R2 run OSPF, routers R2 and R3 run EIGRP (router R2 runs both). R1 advertises a set of networks to R2 via OSPF. R2 redistributes these networks into EIGRP, summarizes them and sends them to R3. R2 wil therefore have both the specific subnets learned from OSPF, and the Null0 discard route corresponding to the summary installed by EIGRP.
Best regards,
Peter
