Task 2.4 IGP Redistribution, route loop?

Hi All,

 

It seems that I am missing something here, because the proposed confiugration in the SG will cause a routing loop. Can someone please comment.

R3 is the ABR for area 38, and R3  will send a default route into area 38 because it is a stub area. But then R1 will redistribute the default route from OSPF into EIGRP and advertise it to R3 via EIGRP,  but R3 was the one who originated the default route and injected in OSPF area 38, so now R3 will install this default route via EIGRP, since R3 being the ABR for area 38 only advertsied it via OSPF area 38 without actually installing it :

 

 

Rack1R1#sh ip route ospf 

     174.1.0.0/16 is variably subnetted, 10 subnets, 2 masks

O       174.1.38.0/24 [110/4] via 174.1.31.9, 01:46:02, FastEthernet0/0.13

O       174.1.34.0/24 [110/2] via 174.1.31.9, 01:46:51, FastEthernet0/0.13

O       174.1.24.0/24 [110/3] via 174.1.31.9, 01:46:22, FastEthernet0/0.13

O IA    174.1.23.3/32 [110/4] via 174.1.31.9, 00:33:58, FastEthernet0/0.13

     150.1.0.0/16 is variably subnetted, 8 subnets, 2 masks

O       150.1.10.10/32 [110/3] via 174.1.31.9, 01:46:22, FastEthernet0/0.13

O       150.1.9.9/32 [110/2] via 174.1.31.9, 01:46:51, FastEthernet0/0.13

O       150.1.8.8/32 [110/4] via 174.1.31.9, 01:46:02, FastEthernet0/0.13

O IA    150.1.3.3/32 [110/5] via 174.1.31.9, 01:46:02, FastEthernet0/0.13

O*IA 0.0.0.0/0 [110/5] via 174.1.31.9, 01:46:02, FastEthernet0/0.13






Now R1 will redisribute it back into the entire EIGRP domain, which is wrong for two reasons:


1- this is actually an invalid route outside area 38,


2- R3 (the OSPF ABR which orignally injected this default route into area 38) will install the route via EIGRP, and this will cause a route loop between R1 ->  R3 -> SW2 -> SW3 -> SW4



Rack1R3#sh ip rout eigrp

D EX 222.22.2.0/24 [170/2561024256] via 174.1.13.1, 02:22:49, Serial1/2

D EX 220.20.3.0/24 [170/2561024256] via 174.1.13.1, 02:22:49, Serial1/2

     174.1.0.0/24 is subnetted, 9 subnets

D       174.1.145.0 [90/2681856] via 174.1.13.1, 04:13:57, Serial1/2

D       174.1.45.0 [90/2684416] via 174.1.13.1, 04:12:57, Serial1/2

D EX 192.10.1.0/24 [170/2561024256] via 174.1.13.1, 02:22:49, Serial1/2

     150.1.0.0/16 is variably subnetted, 7 subnets, 2 masks

D       150.1.5.0/24 [90/2809856] via 174.1.13.1, 04:12:57, Serial1/2

D       150.1.4.0/24 [90/2809856] via 174.1.13.1, 04:12:57, Serial1/2

D       150.1.1.0/24 [90/2297856] via 174.1.13.1, 04:14:20, Serial1/2

D EX 205.90.31.0/24 [170/2561024256] via 174.1.13.1, 02:22:49, Serial1/2

D*EX 0.0.0.0/0 [170/3072256] via 174.1.13.1, 00:00:07, Serial1/2

Rack1R3#

 

 

 

Comments

  • hi ,

    i have also found this very confusing ,

    i think what the Solution is trying to say is that if you dont redistrbute OSPF into EIGRP , one way ONLY on R1 then R4 and R5 wont forward to the external Eigrp routes..

    this is because "EIGRP wont forward prefixs if there not in the routing table" .

    i personaly think its more about the split horizon than the redistribution. and the fact that R1 wont send on the externally learned prefixes to R4 and R5 . 

    but saying that , its very confusing what there trying to get across .!!.

    so even though you would get the loop you showed , that seems to be a side effect of the actual problem.

    Confused ... you bet !

    Hth

  • Hi,

    Can't see the lab now but one question, R1 also belongs to area 38 (which is stub area? and you are redistirbuting into stub area?)

    Waiting for your to add more suggestion.

    Good Luck

  • Hi,

    Can't see the lab now but one question, R1 also belongs to area 38 (which is stub area? and you are redistirbuting into stub area?)

    Waiting for your to add more suggestion.

    Good Luck

     

    I did not redistribute EIGRP into the OSPF stub area, the only redistribution on R1 was OSPF -> EIGRP. And I understand the reason of this redistribution: R1 will prefer the routes for the OSPF area 38 networks since they have a lower administrative distance (110) than the routes for the same networks comming from R3 EIGRP redidstribution (170), thus R1 will not propagate those routes (the area 38 networks) to R4 and R5 via EIGRP.

     

    BTW, I have edited my question to make it more clear, hopefully :)

Sign In or Register to comment.