01-19-2011 02:27 AM - edited 03-04-2019 11:08 AM
Hi,
I seem to be stuck understanding the results on the lab im working on. If you look at the image, i have redistribution configured on R2 and R3 (redistributing EIGRP and OSPF in both directions) and i have RIP being redistributed by R5 into the EIGRP domain. I have made no changes to alter any metrics at all. I have just enabled redistribution.
If I log onto R2 and check the routing table, to get to 50.50.1.0/50.50.2.0/50.50.3.0, it will go through the OSPF domain...next hop being R4.
R2#show ip route
Gateway of last resort is not set
50.0.0.0/24 is subnetted, 4 subnets
O E1 50.50.1.0 [110/22] via 172.41.0.4, 00:38:35, FastEthernet0/0
O E1 50.50.2.0 [110/22] via 172.41.0.4, 00:38:35, FastEthernet0/0
O E1 50.50.3.0 [110/22] via 172.41.0.4, 00:38:35, FastEthernet0/0
O E1 50.50.4.0 [110/22] via 172.41.0.4, 00:38:35, FastEthernet0/0
If i log onto R3 and check the routing rable, to get to 50.50.1.0/50.50.2.0/50.50.3.0, it will go through the EIGRP domain, next hop being R1.
R3#show ip route
Gateway of last resort is not set
50.0.0.0/24 is subnetted, 4 subnets
D EX 50.50.1.0 [170/26117120] via 10.45.31.1, 00:54:20, Serial0/0
D EX 50.50.2.0 [170/26117120] via 10.45.31.1, 00:54:20, Serial0/0
D EX 50.50.3.0 [170/26117120] via 10.45.31.1, 00:54:20, Serial0/0
D EX 50.50.4.0 [170/26117120] via 10.45.31.1, 00:54:20, Serial0/0
I checked the EIGRP topology table (all-links) on R2 and it has the same FD as R3's EIGRP topology table.
R2#show ip eigrp top all-links
IP-EIGRP Topology Table for AS(1)/ID(22.22.22.22)
Codes: P - Passive, A - Active, U - Update, Q - Query, R - Reply,
r - reply Status, s - sia Status
P 50.50.1.0/24, 1 successors, FD is 25602560, serno 14
via Redistributed (25602560/0)
via 10.45.21.1 (26117120/25605120), Serial0/0
R3#show ip eigr top
IP-EIGRP Topology Table for AS(1)/ID(33.33.33.33)
Codes: P - Passive, A - Active, U - Update, Q - Query, R - Reply,
r - reply Status, s - sia Status
P 50.50.1.0/24, 1 successors, FD is 26117120
via 10.45.31.1 (26117120/25605120), Serial0/0
Is there a reason why one router goes one way and the other router goes the other direction (longer path)??
Thanks
DK
Solved! Go to Solution.
01-19-2011 04:26 AM
Hi,
I cleared the IP routing table and ospf process on both R2 and R3 and it swoped the paths around. Now R2 takes the path through EIGRP and R3 now goes backwards through OSPF.
This is normal because of your 2way mutual redistribution between OSPF and EIGRP you have route feedback that is one route redistributed
from one protocol comes back in from the other protocol.
You must deal with administrative distance or tag routes and filter on tag to solve your problem.
take a look here: http://blog.ine.com/2008/02/09/understanding-redistribution-part-i/
Regards.
Alain.
01-19-2011 02:45 AM
Hi Daniel,
It looks to me like this may be do to with the Admin Distance of the learned routes. External EIGRP (redistributred) routes have an AD of 170, whereas the OSPF learned routes have an AD of 110, making them preferable.
It looks like R2 has decided that the OSPF learned route to the 50.50.x.x networks are perferable to the external EIGRP ones it will also see. R3 may well have picked up and installed the direct route to the 50.50.x.x via R1 before OSPF converged properly?
Does clearing the IP ropute tables or restarting the OSPF processes on the R2 & R3 routers cause this behaviour to change?
Cheers
Jonathan
01-19-2011 03:04 AM
Hi Jonathan,
I cleared the IP routing table and ospf process on both R2 and R3 and it swoped the paths around. Now R2 takes the path through EIGRP and R3 now goes backwards through OSPF.
I even deleted all redistribution configurations on R2 and R3, re added them and its still doing the same thing.
01-19-2011 04:26 AM
Hi,
I cleared the IP routing table and ospf process on both R2 and R3 and it swoped the paths around. Now R2 takes the path through EIGRP and R3 now goes backwards through OSPF.
This is normal because of your 2way mutual redistribution between OSPF and EIGRP you have route feedback that is one route redistributed
from one protocol comes back in from the other protocol.
You must deal with administrative distance or tag routes and filter on tag to solve your problem.
take a look here: http://blog.ine.com/2008/02/09/understanding-redistribution-part-i/
Regards.
Alain.
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide