CCIE - Internetwork Expert's Online Community

Hi,

IGP redistribution, after the solution i see that on R1 and R2 OSPF external routes which are resdistributed from RIP seen as EIGRP external route. isn`t that problem?

I'm running on the same problem here. As I see it there are two ways to look at this problem:

- We can leave the sub-optimal routing in place. Connectivity to those networks are not being affected, there just aking a suboptimal path. Optimal path routing is not a requirement for this section.

- We can filter the routes from being redistributed back into EIGRP if they were originated by the same EIGRP process.

Just my two cents.

I chose to filter any routes that originated in EIGRP (or behind them in RIP) from being sent back to EIGRP.  In most cases, I think there's little value in telling a protocol about networks that you learned from it in the first place.  Kind of like a manual split horizon mechanism. 

I figure that if I don't allow routes originated behind OSPF to come in from EIGRP, and if I don't allow EIGRP routes to come back from OSPF, then I should be preventing really poor routing from happening.

On the topic of this initial post, my issue here was that I still needed to make sure that I was changing the distance of external OSPF to >170, as I was not getting proper path information even though I thought the rest of my configuration looked OK.    R1 and R2 both appeared to have some funny things, like this route on R2:

P 31.3.0.0/16, 1 successors, FD is 2560000256, tag is 110
        via Redistributed (2560000256/0)

which was being sent to OSPF rather than via EIGRP up to the RIP domain.  Making the external OSPF distance 171 "fixed" that.  Before I changed the external ospf distance, I was getting spotty results when trying to make sure BB1 routes and BB3 routes exited the network properly.

Is that what was being asked?