Changing EIGRP metrics

[insubliminal]

Hello all,

I have a branch office with a 3560 as its main LAN router. All traffic is told from this router to go out a 2610 over a 1.5MB line. Their is a backup line running off a MPLS 2820 router via a 1.5MB line as well. All 3 of these routers are running EIGRP. My problem is since both lines are 1.5MB, how can I tell the 3560 to prefer the T1 off the 2610 rather than the 2820. I want the 2820 as the failover router. Would I also have to change the metrics at the remote site? Any help would do. Thanks,

Jimmy K

 

[vipergg]

Try here

http://cisco.com/en/US/tech/tk365/technologies_tech_note09186a00800c2d96.shtml

 

[helpdeskdan]

Eigrp supports load balancing - why not just use both?

 

[insubliminal]

Hi Viper,

Thanks for the doc. I tried a few things from it and not getting good results but maybe I am doing something wrong.

Helpdeskdan, if you dont mind maybe giving me an idea on how to configure load balancing? Thanks,

Jimmy K

 

[globalchicken]

you would change the variance with a multiplier that would include both links.

http://www.cisco.com/warp/public/103/19.html#topic1

 

[globalchicken]

Just realized that you said both links were same speed.

Because EIGRP uses the interface bandwidth to determine the rate at which to send packets, it is important that these be set correctly. If it is necessary to influence the path EIGRP chooses, always use delay to do so.

At lower bandwidths, the bandwidth has more influence over the total metric; at higher bandwidths, the delay has more influence over the total metric.

 

[globalchicken]

Also by default EIGRP will load balance across 4 equal cost paths..Below is another excerpt from cisco...

Load Balancing
EIGRP puts up to four routes of equal cost in the routing table, which the router then load-balances. The type of load balancing (per packet or per destination) depends on the type of switching being done in the router. EIGRP, however, can also load-balance over unequal cost links.

Note: Using max-paths, you can configure EIGRP to use up to six routes of equal cost.

Let us say there are four paths to a given destination, and the metrics for these paths are:

path 1: 1100

path 2: 1100

path 3: 2000

path 4: 4000

The router, by default, places traffic on both path 1 and 2. Using EIGRP, you can use the variance command to instruct the router to also place traffic on paths 3 and 4. The variance is a multiplier: traffic will be placed on any link that has a metric less than the best path multiplied by the variance. To load balance over paths 1, 2, and 3, use variance 2, because 1100 x 2 = 2200, which is greater than the metric through path 3. Similarly, to also add path 4, issue variance 4 under the router eigrp command

 

[insubliminal]

Hi globalchicken,

Thanks for your response. Can you give me some sample configs on your proposal above or at least the commands to lead me the right way? Thanks,

Jimmy K

 

[globalchicken]

Here is a link on complete configuration of EIGRP

http://www.cisco.com/univercd/cc/td/doc/product/software/ios122/122cgcr/fipr_c/ipcprt2/1cfeigrp.htm

Its hard to reccomend changes to your configs without knowing what your config is.

I suggest gathering information on your devices concerning your eigrp topology and metrics consult the above link and think about where you want to go from there. Like my post above says, is that by default, identical paths aka metrics, will be automatically load balanced. So figure out what your metrics are between the 2 links and you then have 2 options: 1. Change your metrics to be identical, or use the variance command to include both links. Remember this will load balance the links to prefer one over the other.

To consider one link over the other, use your metrics to change EIGRP to prefer your link.

If you can post as much EIGRP top information and then we can procede to help you out.

 

[insubliminal]

Hello globalchicken or anyone else willing to help,

Here are my configs from my lab environment

2610 (production is the 3560)

version 12.1
no service single-slot-reload-enable
service timestamps debug uptime
service timestamps log uptime
no service password-encryption
!
hostname RouterA
!
no logging rate-limit
enable secret 5 $1$DEcy$m9u.UFc29rmKi95dIsxd//
enable password billy
!
ip subnet-zero
!
!
no ip finger
no ip domain-lookup
!
call rsvp-sync
!
!
!
!
!
!
!
!
interface Loopback0
ip address 1.1.1.1 255.255.255.0
!
interface Ethernet0/0
ip address 192.168.101.254 255.255.255.0
no ip redirects
no keepalive
half-duplex
!
interface Serial0/0
no ip address
no ip redirects
!
router eigrp 50
network 192.168.101.0
no auto-summary
no eigrp log-neighbor-changes
!
no ip classless
ip route 0.0.0.0 0.0.0.0 192.168.101.253

no ip http server
!
access-list 99 permit 192.168.0.0 0.0.255.255
tacacs-server host 192.168.1.203
tacacs-server key 135792468
!
dial-peer cor custom
!
!
!
Next Router
2610 (Production is a 2610)

version 12.1
no service single-slot-reload-enable
service timestamps debug uptime
service timestamps log uptime
no service password-encryption
!
hostname RouterB
!
logging rate-limit console 10 except errors
enable secret 5 $1$5Do.$GOcnjwkLmmo.jpnsi/2nj/
!
ip subnet-zero
!
!
no ip finger
no ip domain-lookup
!
call rsvp-sync
!
!
!
!
!
!
!
!
interface Loopback0
ip address 2.2.2.2 255.255.255.255
!
interface Ethernet0/0
ip address 192.168.101.253 255.255.255.0
no keepalive
half-duplex
!
interface Serial0/0
ip address 192.168.252.22 255.255.255.252
no fair-queue
!
interface Serial0/1
no ip address
!
router eigrp 50
network 192.168.101.0
network 192.168.252.20 0.0.0.3
no auto-summary
no eigrp log-neighbor-changes
!
no ip classless
ip route 0.0.0.0 0.0.0.0 192.168.1.214
no ip http server
!
!
dial-peer cor custom
!
Next Router
3640 (production is ISP 2801 router running MPLS/BGP)

version 12.1
no service single-slot-reload-enable
service timestamps debug uptime
service timestamps log uptime
no service password-encryption
!
hostname RouterC3640
!
logging rate-limit console 10 except errors
enable secret 5 $1$9TjY$q0pe.GCtsvD9YLMwwoVNL1
!
ip subnet-zero
!
!
no ip finger
no ip domain-lookup
!
call rsvp-sync
cns event-service server
!

!
!
controller T1 1/0
!
controller T1 1/1
!
!
interface Loopback0
ip address 3.3.3.3 255.255.255.255
!
interface FastEthernet0/0
ip address 12.12.12.17 255.255.255.252
no keepalive
duplex auto
speed auto
!
interface Serial0/0
ip address 12.6.6.29 255.255.255.252
shutdown
service-module t1 clock source internal
!
interface Serial0/1
ip address 12.8.8.10 255.255.255.252
shutdown
clockrate 768000
!
interface FastEthernet1/0
ip address 192.168.101.1 255.255.255.0
no ip redirects
duplex auto
speed auto
!
interface FastEthernet1/1
ip address 172.16.101.1 255.255.255.0
no ip redirects
duplex auto
speed auto
!
router eigrp 50
redistribute bgp 65009 metric 10000 10 255 1 1500
network 172.16.0.0
network 192.168.101.0
no auto-summary
no eigrp log-neighbor-changes
!
router bgp 65009
no synchronization
no bgp log-neighbor-changes
network 172.16.101.0 mask 255.255.255.0
network 192.168.101.0
neighbor 12.12.12.18 remote-as 65000
no auto-summary
!
ip kerberos source-interface any
ip classless
ip route 0.0.0.0 0.0.0.0 FastEthernet0/0
no ip http server
!
access-list 99 permit 192.168.0.0 0.0.255.255
!
!
dial-peer cor custom
!

These are the configs for the 3 test routers. the 2610 (Local LAN RouterA) has 2 directions to go. In this test environment, an ethernet interface is back to back with RouterC (3640) from Router A and DCE/DTE cable from Router A to Router B. Of course in this test environment, the preferred router will be the ethernet connection. This will be irrelevant in the test as I want to force EIGRP traffic towards RouterB.
In Production, they are both T1's.
Let me know if you need any more configs. What I have tried to do is change metrics, change delay, add static routers, and for some reason the routes learned form EIGRP vie redistribtion from BGP always precede even with a AD of External EIGRP 170.

Here is a sh ip route currently of RouterA

RouterA#sh ip route
Codes: C - connected, S - static, I - IGRP, 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, E - EGP
i - IS-IS, 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 route

Gateway of last resort is 192.168.101.253 to network 0.0.0.0

1.0.0.0/24 is subnetted, 1 subnets
C 1.1.1.0 is directly connected, Loopback0
172.16.0.0/24 is subnetted, 1 subnets
D 172.16.101.0 [90/284160] via 192.168.101.1, 00:13:59, Ethernet0/0
D 192.168.0.0/24 [90/2221312] via 192.168.101.253, 00:14:02, Ethernet0/0
12.0.0.0/30 is subnetted, 1 subnets
D 12.7.7.4 [90/2733056] via 192.168.101.253, 00:14:02, Ethernet0/0
D 192.168.1.0/24 [90/2221056] via 192.168.101.253, 00:14:02, Ethernet0/0
192.168.252.0/30 is subnetted, 1 subnets
D 192.168.252.20 [90/2195456] via 192.168.101.253, 00:14:03, Ethernet0/0
D EX 192.168.3.0/24 [170/284160] via 192.168.101.1, 00:14:29, Ethernet0/0
C 192.168.101.0/24 is directly connected, Ethernet0/0
S* 0.0.0.0/0 [1/0] via 192.168.101.253
RouterA#

No matter what I change, the 192.168.3.0 route always is preferred with AD 170. This is traffic I want to force out RouterB via 192.168.101.253.

Keep in mind these are edge router configs and all routers eventually lead to the same core. Any help will do.

Thanks,

Jimmy K

 

[globalchicken]

Why run EIGRP on router A? You could implement a static route that points to your primary link and then a backup floating static route that points to your redundant link.

ip route 0.0.0.0 primary_ip 5
ip route 0.0.0.0 secondary_ip 15

If all traffic must pass on to your edge router then point everything to it. If you must run eigrp to communicate with internal routers, use a different process id and do not redistribute that to your edge routers.

 

[insubliminal]

Hi global,

I tried using those static routers and networks destined for the 192.168.0.0 network such as the 3.0 network still go out RouterC whereas I want to force it to Router B.

I put in
ip route 0.0.0.0 0.0.0.0 192.168.101.253 5
ip route 0.0.0.0 0.0.0.0 192.168.101.1 15

and packets destined for the 192.168.3.0 network still go abide the EIGRP ecternal AD of 170. I am confused. Now if I just put a static route of
ip route 192.168.3.0 255.255.255.0 192.168.101.253 5 then yes, it takes in this case since I guess it is more specific and then sends it out RouterB, but then I do not have any dynamic routing as I added a secondary route
ip route 192.168.3.0 255.255.255.0 192.168.101.1 15 and when I shut RouterB's e0/0 interface, the traffic does not know to go out the other way via the secondary weighted route.

Jimmy K

 

[globalchicken]

Putting in your static route of
0.0.0.0 only sends traffic that way if it does not have any other option (gateway of last resort) You already have ex EIGRP saying where 192.168.3.0 should go. Adding that route with EIGRP functioning will not send traffic there because it already has a route. AM i making sense?

Another thing i dont understand about your topology is why are you redistributing bgp back into your internal router? He does not need to know about these external routes or does he? Not sure why you are doing it that way.

You could put in just the one static route

ip route 192.168.3.0 255.255.255.0 192.168.101.253 5

and then use your external eigrp to put the route to your redundant router back into the routing table... I dont know from the information above why your second static route did not work when your routerB interface went down. With a administrative distance of 15 this should be much more feasible to that network than any other route is.

Hmm

M

 

[helpdeskdan]

Good point, globalchicken. You can mess with AD till you are blue in the face, but if there is a more exact route in the routing table, (longest subnet mask) it will pick that route, no matter what the AD. In fact, it won't use the default route unless there is no other route AT ALL.