Hi Rene
I got stuck in doing this. Can some one let me know the configuration for this
Thanks
Peter
RIP EIGRP Redistribution Route Tagging
Scenario:
The Australia Zoo has hired you as a crocodile hunter network engineer responsible for all bits & bytes in the zoo. As an experiment the zoo decided to have the kangaroos and dolphins around the same lake to study the interaction between the animals. Both departments need to have access to each other's data which means you'll have to exchange routing information. To make sure there is no single point of failure you will use router Lassie and Willy to configure two-way redistribution. This however will also introduce new problems like routing loops or sub-optimal routing, it's up to you to enable redistribution and solve any problems along the way...
Goal:
- All IP addresses have been preconfigured for you.
- RIP and EIGRP have been preconfigred for you on the corresponding routers.
- Enable two-way redistribution between RIP and EIGRP on router Lassie and Willy.
- You notice that router Willy or Lassie are sending traffic to 4.4.4.4 towards router Flipper, make sure you get rid of this sub-optimal routing.
- Use route tagging to accomplish this.
IOS:
c3640-jk9s-mz.124-16.bin
Topology:
You need to register to be able to download the GNS3 Topology File. (Registration is Free!)
- Related Articles
Only registered users can write comments!
Comments (23)
-
peter_mlr 2011-03-05 17:26:00
-
Mehdi93$ 2011-03-13 14:55:31I got stuck in doing this as well. Can some one let the final configuration.
Thanks in advance.
Mehdi.
-
ReneMolenaar 2011-03-14 15:28:17Hi Guys,
I assume you know how to do redistribution so i'll keep it short to the route tagging part.
In this scenario we have "two-way redistribution" and we want to make sure prefixes from one routing protocol are not advertised again into the same routing protocol.
Example: network 4.4.4.0 will be redistributed INTO RIP at router Lassie. The redistributed network 4.4.4.0 will be redistributed AGAIN into EIGRP at router Willy.
If we put a tag on the redistributed prefixes than we can make sure they won't be redistributed again.
Here we go:
When you configure redistribution you specify a route-map.
router rip
redistribute eigrp 234 metric 2 route-map EIGRP2RIProuter eigrp 234
redistribute rip route-map RIP2EIGRP metric 1 1 1 1 1When we find a prefix that is tagged with "88" we don't redistribute it
route-map EIGRP2RIP deny 10
match tag 88When we redistribute we tag the prefix with "77".
route-map EIGRP2RIP permit 20
set tag 77route-map RIP2EIGRP deny 10
match tag 77route-map RIP2EIGRP permit 20
set tag 88So if we see something that is tagged we don't redistribute it, otherwise we tag it and redistribute it.
Does this make sense? good luck!
Rene
-
Mehdi93$ 2011-03-16 01:23:51
ok thank you that's great !;D
Mehdi.
-
kenjidnb 2011-03-23 02:07:45I have configured 2-way redistribution without route maps but I don't see the routing loops (all my pings are successful from skippy to flipper and from flipper to skippy).
My config on Lassie (same redistribution commands on Willie):
router eigrp 234
redistribute rip metric 1 1 1 1 1
network 192.168.24.0
no auto-summary
!
router rip
version 2
redistribute eigrp 234 metric 2
network 192.168.12.0
no auto-summary
And routing tables:
Skippy#sh ip route | b Gate
Gateway of last resort is not setD EX 192.168.12.0/24
[170/2560002816] via 192.168.34.3, 00
2:10, FastEthernet1/0
[170/2560002816] via 192.168.24.2, 002:10, FastEthernet0/0
1.0.0.0/24 is subnetted, 1 subnets
D EX 1.1.1.0 [170/2560002816] via 192.168.34.3, 002:10, FastEthernet1/0
[170/2560002816] via 192.168.24.2, 002:10, FastEthernet0/0
D EX 192.168.13.0/24
[170/2560002816] via 192.168.34.3, 002:10, FastEthernet1/0
[170/2560002816] via 192.168.24.2, 002:10, FastEthernet0/0
4.0.0.0/24 is subnetted, 1 subnets
C 4.4.4.0 is directly connected, Loopback0
C 192.168.24.0/24 is directly connected, FastEthernet0/0
C 192.168.34.0/24 is directly connected, FastEthernet1/0Flipper#sh ip route | b Gate
Gateway of last resort is not setC 192.168.12.0/24 is directly connected, FastEthernet0/0
1.0.0.0/24 is subnetted, 1 subnets
C 1.1.1.0 is directly connected, Loopback0
C 192.168.13.0/24 is directly connected, FastEthernet1/0
4.0.0.0/24 is subnetted, 1 subnets
R 4.4.4.0 [120/2] via 192.168.13.3, 000:20, FastEthernet1/0
[120/2] via 192.168.12.2, 0002, FastEthernet0/0
R 192.168.24.0/24 [120/2] via 192.168.13.3, 000:20, FastEthernet1/0
[120/2] via 192.168.12.2, 0002, FastEthernet0/0
R 192.168.34.0/24 [120/2] via 192.168.13.3, 000:20, FastEthernet1/0
[120/2] via 192.168.12.2, 0002, FastEtherWilly#sh ip route | b Gate
Gateway of last resort is not setR 192.168.12.0/24 [120/1] via 192.168.13.1, 00
02, FastEthernet0/0
1.0.0.0/24 is subnetted, 1 subnets
R 1.1.1.0 [120/1] via 192.168.13.1, 0002, FastEthernet0/0
C 192.168.13.0/24 is directly connected, FastEthernet0/0
4.0.0.0/24 is subnetted, 1 subnets
D 4.4.4.0 [90/156160] via 192.168.34.4, 002:40, FastEthernet1/0
D 192.168.24.0/24 [90/30720] via 192.168.34.4, 002:40, FastEthernet1/0
C 192.168.34.0/24 is directly connected, FastEthernet1/0Lassie#sh ip route | b Gate
Gateway of last resort is not setC 192.168.12.0/24 is directly connected, FastEthernet0/0
1.0.0.0/24 is subnetted, 1 subnets
R 1.1.1.0 [120/1] via 192.168.12.1, 000:10, FastEthernet0/0
R 192.168.13.0/24 [120/1] via 192.168.12.1, 000:10, FastEthernet0/0
4.0.0.0/24 is subnetted, 1 subnets
D 4.4.4.0 [90/156160] via 192.168.24.4, 01:184, FastEthernet1/0
C 192.168.24.0/24 is directly connected, FastEthernet1/0
D 192.168.34.0/24 [90/30720] via 192.168.24.4, 01:184, FastEthernet1/0
Beside suboptimal paths for the 192.168.X.Y (load-balanced equal cost paths), I don't see very bad things in my routing tables. Am I missing something?
-
lela 2011-03-24 18:24:33Hy tulum
Ping is irelevant for loops , you should do a traceroute ,if you will see more than 2 hops .. is a loop
skippi#traceroute 1.1.1.1Type escape sequence to abort.
Tracing the route to 4.4.4.41 192.168.34.3 52 msec
192.168.24.2 28 msec
192.168.34.3 36 msec
2 192.168.12.1 68 msec
192.168.13.1 76 msec
192.168.12.1 52 msec
this is a loop.Generic LOOP PREVENTION solution by tagging routes without using ACL. This is a more scalable approach than ACL based one. It is a one-for-all solution regardless of the numbers of non-distinctive routes.
First Mutual Redistribution Point:
router A #
router X
redistribute Y route-map Y_to_X metric x...router Y
redistribute X route-map X_to_Y metric x...route-map Y_to_X deny 10
match tag AD_Xroute-map Y_to_X permit 20
set tag AD_Yroute-map X_to_Y deny 10
match tag AD_Yroute-map X_to_Y permit 20
set tag AD_XExample
router rip
redistribute eigrp 1 route-map eigrp_to_rip metric 2router eigrp 1
redistribute rip route-map rip_to_eigrp metric 10000 100 255 1 1500route-map eigrp_to_rip deny 10
match tag 120
-
lela 2011-03-24 18:29:59
For some reason previous post was not complete .
This should be the solution NOT found by me i was using access list
First Mutual Redistribution Point:
router A #
router X
redistribute Y route-map Y_to_X metric x...router Y
redistribute X route-map X_to_Y metric x...route-map Y_to_X deny 10
match tag AD_Xroute-map Y_to_X permit 20
set tag AD_Yroute-map X_to_Y deny 10
match tag AD_Yroute-map X_to_Y permit 20
set tag AD_XExample
router rip
redistribute eigrp 1 route-map eigrp_to_rip metric 2router eigrp 1
redistribute rip route-map rip_to_eigrp metric 10000 100 255 1 1500route-map eigrp_to_rip deny 10
match tag 120
-
kenjidnb 2011-03-24 18:51:50
Thanks Ano. I had noticed the extra hops with traceroute too but I assumed it was okay since pings were successful. Now I know that routing loops doesn't necessary mean packet loss.
-
lela 2011-03-24 19:01:45 i was fighting like a ninja with ROUTE REDISTRIBUTION and PATH SELECTION the concept it seems to be simple, but in practice is tricky.
This is a real great site for GNS labs >> cisco stuff
-
lela 2011-03-24 19:05:18
For the 3 time i uplod hope now will be all
Example
router rip
redistribute eigrp 1 route-map eigrp_to_rip metric 2router eigrp 1
redistribute rip route-map rip_to_eigrp metric 10000 100 255 1 1500route-map eigrp_to_rip deny 10
match tag 120
-
ReneMolenaar 2011-03-27 14:26:03
Besides routing loops (you can spot those with traceroute) you can have blackhole routing...test some things and see what happens when neighbors go down and come back online.
-
djtjlt 2011-04-20 13:16:34written by Rene Molenaar, March 14, 2011
Example: network 4.4.4.0 will be redistributed INTO RIP at router Lassie. The redistributed network 4.4.4.0 will be redistributed AGAIN into EIGRP at router Willy.I believe, that there is no suboptimal routing int this lab, as the example is wrong from the beginning, as EIGRP is using different AD for internal and external routes. So following is happening next:
1) When 4.4.4.0 is being redistributed to rip at Lassie, it's AD is 120 (RIP); then when redistributting it back to eigrp at willy, it will have AD 170 (EIGRP external).
2) So router Willy will not use this route (AD 170), neither RIP route (AD 120), because it hears the route for 4.4.4.0 network from Router skippy with AD 90 and put's this route to it's routing table.Eigrp's default AD's 90 and 170 solve this problem automatically.
I think the lab should have been made with two domains - OSPF and RIP.
As ospf internal and external routes both have AD 110, the suboptimal routing can occur.If to take the same example, then if route is being redistributed from ospf domain to rip at lessie - it will get AD 120, and when redistributing back to ospf, it will get AD 110. So router will hear about this route from two sources with the same AD, but different metric. And if metric received when reredistributing from RIP is better than metric, to reach this network directly via skippy router , then router will put this suboptimal route to it's routing table. To avoid this, you can change either AD for external ospf routes, or you can change the metric. To prove this suboptimal route at willy you could launcht trace to 4.4.4.4 and it would show that the routing is going willy -> flipper -> lassie -> skipper instead of directly going to skipper.
And I think that folowing trace shows not the suboptimal routing, but just a equal cost load balance...
1 192.168.34.3 52 msec
192.168.24.2 28 msec
192.168.34.3 36 msec
2 192.168.12.1 68 msec
192.168.13.1 76 msec
192.168.12.1 52 msecCorrect me if I'm wrong
-
ReneMolenaar 2011-04-21 16:43:52
Hi Tadas,
You are right I should have designed it differently so I changed it
EIGRP makes a difference between internal or external which makes sure that internal prefixes are always preferred.
You could still have a problem in the RIP domain since RIP makes no difference between internal or external prefixes.
Let's say that you have an internal RIP prefix with a hop count of 10, and you redistribute this prefix back into RIP with a hop count of 1. You'll get sub-optimal routing.
To demonstrate this I changed router Skippy so the 4.4.4.0 prefix is advertised in EIGRP as an external prefix. You'll see that either router Willy or Lassie will take an sub-optimal path towards 4.4.4.0.
Enjoy and thanks for the comment
;D
-
hameidi 2011-07-28 18:55:07Dear:
kindly add video for this lab
-
uscallesen 2011-09-13 23:40:10@ muhaned - I believe the solution to the sub-optimal routing is to adjust the administrative distance for External EIGRP routes on Router Lassie and Willy to < 120 & > 90 - I used 95. The routes maps are standard as already suggested.
Router Willy (basicly same setup on Lassie)
!
router eigrp 234
redistribute rip metric 100 100 100 100 100 route-map INTO-EIGRP
network 192.168.34.0
distance eigrp 90 95
no auto-summary
!
router rip
version 2
redistribute eigrp 234 metric 5 route-map INTO-RIP
network 192.168.13.0
no auto-summary
!
route-map INTO-EIGRP deny 10
match tag 2
!
route-map INTO-EIGRP permit 20
set tag 1
!
route-map INTO-RIP deny 10
match tag 1
!
route-map INTO-RIP permit 20
set tag 2
!
-
plague22 2011-09-16 20:25:34I'm not sure route tagging is the issue for this lab.
The issue seems to be when you redistribute originally from rip into eigrp you specify a bandwidth when doing so. i.e.r
redistribute rip metric 1500 1 1 100 1500
So from router skippy you will have two paths to reach network 1.1.1.0
D EX 1.1.1.0 [170/1709312] via 192.168.34.3, 000:10, FastEthernet1/0
[170/1709312] via 192.168.24.2, 000:10, FastEthernet0/0
So when you perform a trace you will see more than two hops as its load balancing between the two
Tracing the route to 1.1.1.11 192.168.34.3 32 msec
192.168.24.2 16 msec
192.168.34.3 8 msec
2 192.168.12.1 48 msec
192.168.13.1 24 msec *This isnt a routing loop. This is a result of load balancing down the two links.
If you change the metric on one of the redistributing router for ex on Lassie:
redistribute rip metric 1 1 1 1 1
Lassie(config-router)#You will now only have 1 path for the 1.1.1.0 network
D EX 1.1.1.0 [170/1709312] via 192.168.34.3, 000:26, FastEthernet1/0Final trace is
Tracing the route to 1.1.1.11 192.168.34.3 24 msec 20 msec 8 msec
2 192.168.13.1 40 msec * 24 msecSo in this lab is the design that it should be using both links to reach 1.1.1.0?
I dont see how the tagging would fix this.
-
ibrahim 2011-11-10 15:49:29i just didn't got the idea...
the basic principle for redistribution between RIP and EIGRP say that something like that"domain-loop" shouldn't happen, because both routers "R2 & R3" know that there is a route
for the subnet 4.4.4.4 learned through EIGRP with AD = 90, so when they see that there is
another route coming from the RIP network with AD=120, they are not going to make
redistribution to avoid such loop.
my result was as i said, there was no loop even without route tag configured, so where is the problem?
Lassie(config-router)#do sh ip rou
Codes: C - connected, S - static, R - RIP, M - mobile, B - BGP
D - EIGRP, EX - EIGRP external, O - OSPF, IA - OSPF inter area
N1 - OSPF NSSA external type 1, N2 - OSPF NSSA external type 2
E1 - OSPF external type 1, E2 - OSPF external type 2
i - IS-IS, su - IS-IS summary, L1 - IS-IS level-1, L2 - IS-IS level-2
ia - IS-IS inter area, * - candidate default, U - per-user static route
o - ODR, P - periodic downloaded static routeGateway of last resort is not set
C 192.168.12.0/24 is directly connected, FastEthernet0/0
1.0.0.0/24 is subnetted, 1 subnets
R 1.1.1.0 [120/1] via 192.168.12.1, 000:26, FastEthernet0/0
R 192.168.13.0/24 [120/1] via 192.168.12.1, 000:26, FastEthernet0/0
2.0.0.0/32 is subnetted, 1 subnets
C 2.2.2.2 is directly connected, Loopback0
3.0.0.0/32 is subnetted, 1 subnets
D 3.3.3.3 [90/158720] via 192.168.24.4, 00:35:36, FastEthernet1/0
4.0.0.0/24 is subnetted, 1 subnets
D 4.4.4.0 [90/156160] via 192.168.24.4, 00:36:22, FastEthernet1/0
C 192.168.24.0/24 is directly connected, FastEthernet1/0
D 192.168.34.0/24 [90/30720] via 192.168.24.4, 00:46:10, FastEthernet1/0************************************************************************************
Willy(config-router)#do sh ip rou
Codes: C - connected, S - static, R - RIP, M - mobile, B - BGP
D - EIGRP, EX - EIGRP external, O - OSPF, IA - OSPF inter area
N1 - OSPF NSSA external type 1, N2 - OSPF NSSA external type 2
E1 - OSPF external type 1, E2 - OSPF external type 2
i - IS-IS, su - IS-IS summary, L1 - IS-IS level-1, L2 - IS-IS level-2
ia - IS-IS inter area, * - candidate default, U - per-user static route
o - ODR, P - periodic downloaded static routeGateway of last resort is not set
R 192.168.12.0/24 [120/1] via 192.168.13.1, 00
0:20, FastEthernet0/0
1.0.0.0/24 is subnetted, 1 subnets
R 1.1.1.0 [120/1] via 192.168.13.1, 000:20, FastEthernet0/0
C 192.168.13.0/24 is directly connected, FastEthernet0/0
2.0.0.0/32 is subnetted, 1 subnets
D 2.2.2.2 [90/158720] via 192.168.34.4, 00:35:58, FastEthernet1/0
3.0.0.0/32 is subnetted, 1 subnets
C 3.3.3.3 is directly connected, Loopback0
4.0.0.0/24 is subnetted, 1 subnets
D 4.4.4.0 [90/156160] via 192.168.34.4, 00:36:59, FastEthernet1/0
D 192.168.24.0/24 [90/30720] via 192.168.34.4, 00:46:40, FastEthernet1/0
C 192.168.34.0/24 is directly connected, FastEthernet1/0
could that be the IOS version?Cisco IOS Software, 3600 Software (C3640-JK9S-M), Version 12.4(16), RELEASE SOFTWARE (fc1)
Technical Support: http://www.cisco.com/techsupport
Copyright (c) 1986-2007 by Cisco Systems, Inc.
-
ibrahim 2011-11-10 15:50:33
but thanks anyway!
-
ReneMolenaar 2011-11-10 16:05:31
Hello Ibrahim,
The loopback interface on router Skippy is not advertised in EIGRP but redistributed. The AD of this prefix will be 170 (external) not 90 (internal).What happens is that router Lassie and Willy will have 4.4.4.0 /24 with AD 170 in their routing table. Both are configured to redistribute 4.4.4.0 /24 from EIGRP into RIP.
One of the routers will receive the 4.4.4.0 /24 network through RIP with an AD of 120.
AD 120 is lower (and thus better) than AD 170 so the EIGRP external prefix will be removed from the routing table and the RIP prefix will be installed.
Result: sub-optimal routing (do a traceroute to find this).
-
stemrikar 2012-04-16 10:21:04Hi All,
1st of all I thought that no matter what, there will be suboptimal routing, looking from either direction.
then I thought there has to be a way out.
& here it is acc to me.
Lassie#sh run | s rip|eigrp|map
router eigrp 234
redistribute rip metric 100000 10 255 1 1500 route-map r2e
network 192.168.24.0
distribute-list route-map tag170 in
no auto-summary
router rip
version 2
redistribute eigrp 234 metric 5 route-map e2r
network 192.168.12.0
distance 180
no auto-summary
route-map r2e deny 5
match tag 5
route-map r2e permit 10
set tag 170
route-map e2r deny 5
match tag 170
route-map e2r permit 10
set tag 5
route-map tag170 deny 5
match tag 170
route-map tag170 permit 10
Lassie#(Likewise for Willy)
I have tested it by rebooting the routers and the path remains optimal.
Lassie#tr 4.4.4.4
Type escape sequence to abort.
Tracing the route to 4.4.4.41 192.168.24.4 36 msec * 68 msec
Lassie#
Lassie#tr 1.1.1.1Type escape sequence to abort.
Tracing the route to 1.1.1.11 192.168.12.1 16 msec * 28 msec
Lassie#
-
stemrikar 2012-04-16 10:39:21
contd...
Lassie#tr 1.1.1.1
Type escape sequence to abort.
Tracing the route to 1.1.1.11 192.168.12.1 16 msec * 28 msec
Lassie#Rene, Thanx 4 such good, challenging labs.
Regards
Shoeb
-
ReneMolenaar 2012-04-19 09:06:08
Hi Shoeb,
That looks good to me...redistributing and using route-maps to influence the AD and metric is a good method of dealing with this.
Rene
-
mistborn_soother 2012-05-06 22:20:11I'm stuck with this. I did create route-maps on Willy and Lassie as follows
Willy#show route-map
route-map EIGRP_TO_RIP, permit, sequence 10
Match clauses:
Set clauses:
tag 77
Policy routing matches: 0 packets, 0 bytes
route-map EIGRP_TO_RIP, deny, sequence 20
Match clauses:
tag 88
Set clauses:
Policy routing matches: 0 packets, 0 bytes
route-map RIP_TO_EIGRP, deny, sequence 10
Match clauses:
tag 77
Set clauses:
Policy routing matches: 0 packets, 0 bytes
route-map RIP_TO_EIGRP, permit, sequence 20
Match clauses:
Set clauses:
tag 88
Policy routing matches: 0 packets, 0 bytesLassie
route-map EIGRP_TO_RIP, permit, sequence 10
Match clauses:
Set clauses:
tag 77
Policy routing matches: 0 packets, 0 bytes
route-map EIGRP_TO_RIP, deny, sequence 20
Match clauses:
tag 88
Set clauses:
Policy routing matches: 0 packets, 0 bytes
route-map RIP_TO_EIGRP, deny, sequence 10
Match clauses:
tag 77
Set clauses:
Policy routing matches: 0 packets, 0 bytes
route-map RIP_TO_EIGRP, permit, sequence 20
Match clauses:
Set clauses:
tag 88
Policy routing matches: 0 packets, 0 bytes
but although now the ASBRs send direct to Skippy for the 4.4.4.4 route, Flipper's routing table looks like thisC 192.168.12.0/24 is directly connected, FastEthernet0/0
1.0.0.0/24 is subnetted, 1 subnets
C 1.1.1.0 is directly connected, Loopback0
C 192.168.13.0/24 is directly connected, FastEthernet1/0
4.0.0.0/24 is subnetted, 1 subnets
R 4.4.4.0 [120/10] via 192.168.12.2, 0000, FastEthernet0/0
R 192.168.24.0/24 [120/10] via 192.168.12.2, 0000, FastEthernet0/0
R 192.168.34.0/24 [120/10] via 192.168.12.2, 0000, FastEthernet0/0
surely Flipper should have two paths to 4.4.4.4. What happened to via 192.168.13.0?