CCIE - Internetwork Expert's Online Community

Hi,

The PPP link between R4 and R5 will be a backup of both the Ethernet and Frame Relay segments between them.
Both R4 and R5 should be able to detect a failure of either the Frame Relay or Ethernet segments within 10 seconds.
In the case that both the Frame Relay and the Ethernet segmentsbetween R4 and R5 are down traffic should be sent over the PPP link.

the solution for them :

interface Ethernet0/0
 ip ospf hello-interval 2
!
interface Serial1/0
 ip ospf hello-interval 2

That is OK.

But for this one:

•  As long as either the Frame Relay or the Ethernet segment is up no traffic
should be sent over the PPP link

On previous one, 3 intarfaces are load balancing. And with changing hello interval of another interface you can not change the cost. After chaning timers PPP link is still forwarding traffic.

Rack1R4(config-if)#do sh ip ospf int bri
Interface    PID   Area            IP Address/Mask    Cost  State Nbrs F/C
Lo0          1     2               150.1.4.4/24       1     LOOP  0/0
Se1/1        1     2               141.1.45.4/24      1     P2P   1/1
Se1/0        1     2               141.1.54.4/24      1     P2P   1/1
Fa0/0        1     2               141.1.145.4/24     1     BDR   1/1

akyuznet:

On previous one, 3 intarfaces are load balancing. And with changing hello interval of another interface you can not change the cost. After chaning timers PPP link is still forwarding traffic.

If you've gotten your answer already, I apologize.

In tast 4.6, we were asked to load balance between the Frame-Relay and Ethernet connections between R4 and R5.  This didn't include the PPP link, so we shouldn't be load balancing across all 3 links, just 2 of them.  In updating our hello timers on the Frame and Ethernet, we detect the failure more quickly, and if they both fail, the PPP link will be there to take over (at a higher cost, based on interface bandwidth).

Make sense?

--JB

Read carefully the task load balancing are between F.R interface and Fastethernet interface.

Rgds.

hi akyuznet,

this might be a typo, but in your post it looks like you might have accidentally changed the bandwidth of your p2p interface s1/1 because the cost reflects 1 for FastEthernet. it should be 64. this would ensure that load balancing is only performed on equal cost interfaces, your f0/0 which is 100000 and your frame relay s1/0 interface which in this solution, you should change to 100000 to match your FastEthernet interface. however, the output you pasted shows two different type of Ethernet interfaces, the first is a Ethernet 10 Mbps, while the second is a FastEthernet. not sure which one is the one you are using in your config, but if it is Ethernet, then both frame interface s1/0 and eth0 will have a bandwidth of 10000.

jorge