cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
437
Views
0
Helpful
2
Replies

Redistribution problem

moses12315
Level 1
Level 1

We have two networks , until now they were not connected. One of the Network (call A) is running Rip (version 1). The other (call it B)

is running EIGRP. The last device of network B is a 4500 series Switch. The last device of network A is a 2600 series Router.

I connected those devices with a cat6 cable. On the 4500 series Switch i configure "no swithport" and put an ip address. On the F0/0

of the router an ip address of the same network. Both devices are communicate.

I run router Rip version 1 on 4500 switch , and i put that network under Rip . Then i redistribute on EIGRP the Rip routing protocol and on Rip the EIGRP .

When i configured under Rip the EIGRP i put default - metric 2 .

I realised that from other devices in B network(from 4500 switch no problem existed) the communication had a problem to go to network A from network B(loosing 60% of the packets). Then i put default-metric 1 and the problem was solved.

Can anyone explain me what actually that metric is doing? What's its purpose? Why i had that problem with metric 2?

Will i have the same problem if i configure the port of the switch as a swithport ?

Thanks a lot

Moses

2 Accepted Solutions

Accepted Solutions

Jon Marshall
Hall of Fame
Hall of Fame

moses12315 wrote:

We have two networks , until now they were not connected. One of the Network (call A) is running Rip (version 1). The other (call it B)

is running EIGRP. The last device of network B is a 4500 series Switch. The last device of network A is a 2600 series Router.

I connected those devices with a cat6 cable. On the 4500 series Switch i configure "no swithport" and put an ip address. On the F0/0

of the router an ip address of the same network. Both devices are communicate.

I run router Rip version 1 on 4500 switch , and i put that network under Rip . Then i redistribute on EIGRP the Rip routing protocol and on Rip the EIGRP .

When i configured under Rip the EIGRP i put default - metric 2 .

I realised that from other devices in B network(from 4500 switch no problem existed) the communication had a problem to go to network A from network B(loosing 60% of the packets). Then i put default-metric 1 and the problem was solved.

Can anyone explain me what actually that metric is doing? What's its purpose? Why i had that problem with metric 2?

Will i have the same problem if i configure the port of the switch as a swithport ?

Thanks a lot

Moses

Moses

For RIP the metric is hop count so that is what you were specifying. Each time a the traffic goes through a router the hop count is incremented by 1. A hop count of 16 is considered an unreachable network ie. you can't have a route with more than 15 hops. So it may be that by using 2 you actually managed to get to a hop count of 16 for some of the routes thus making them unreachable.

This is one of the reason why RIP is only suited to very small networks.

Jon

View solution in original post

Giuseppe Larosa
Hall of Fame
Hall of Fame

Hello Moses,

Jon is right the reason for the issues may be that you have a daisy chain topology and devices far from the redistributing device may consider some routes as useless.

if so this means that your network is almost too big for RIP.

to be noted that Ripv2 has improvements but not on this aspect.

Hope to help

Giuseppe

View solution in original post

2 Replies 2

Jon Marshall
Hall of Fame
Hall of Fame

moses12315 wrote:

We have two networks , until now they were not connected. One of the Network (call A) is running Rip (version 1). The other (call it B)

is running EIGRP. The last device of network B is a 4500 series Switch. The last device of network A is a 2600 series Router.

I connected those devices with a cat6 cable. On the 4500 series Switch i configure "no swithport" and put an ip address. On the F0/0

of the router an ip address of the same network. Both devices are communicate.

I run router Rip version 1 on 4500 switch , and i put that network under Rip . Then i redistribute on EIGRP the Rip routing protocol and on Rip the EIGRP .

When i configured under Rip the EIGRP i put default - metric 2 .

I realised that from other devices in B network(from 4500 switch no problem existed) the communication had a problem to go to network A from network B(loosing 60% of the packets). Then i put default-metric 1 and the problem was solved.

Can anyone explain me what actually that metric is doing? What's its purpose? Why i had that problem with metric 2?

Will i have the same problem if i configure the port of the switch as a swithport ?

Thanks a lot

Moses

Moses

For RIP the metric is hop count so that is what you were specifying. Each time a the traffic goes through a router the hop count is incremented by 1. A hop count of 16 is considered an unreachable network ie. you can't have a route with more than 15 hops. So it may be that by using 2 you actually managed to get to a hop count of 16 for some of the routes thus making them unreachable.

This is one of the reason why RIP is only suited to very small networks.

Jon

Giuseppe Larosa
Hall of Fame
Hall of Fame

Hello Moses,

Jon is right the reason for the issues may be that you have a daisy chain topology and devices far from the redistributing device may consider some routes as useless.

if so this means that your network is almost too big for RIP.

to be noted that Ripv2 has improvements but not on this aspect.

Hope to help

Giuseppe

Review Cisco Networking products for a $25 gift card