RIP Offset-lists

Lessons Learned:
RIP Metric Calculation

RIP uses hop-count as it’s metric

-1 hop per interface

-16 is infinite

Offset-list – use to modify the metric

-Access-list 0 means all routes

-----------------------------------------------------

Rip uses a hop count as it’s metric and 16 is the maximum hop counts that RIP will support.

To modify this we can use offset lists. Offset lists can be used for all routes or on a prefix basis.

The list can be either inbound or outbound

RIP – the default process for updates, routers will increment the metric of routes as the prefix is advertised out its interface, NOT as the prefix is received.

We can manually increment the hop count / metric as the update is received by using an offset-list.

Also, lists and be used for traffic engineering or filtering.

Note: routes that have a metric of 16 or higher will not be installed in the routing table.

------------------------------------------------------

R2(config-router)#offset-list 0 out 10 fastEthernet 0/1
This is applied under the Rip process. The list # 0 indicated ALL networks. The list is in the outbound direction out fa0/1.

Once the network convergences. The path selection will change.

We can force the faster change by clearing the route table
R2#clear ip route *

Note: when changing the metric you could run into a transient loop scenario where the old routing information has not yet cleared itself from the table.

R2#sh run | b router rip

router rip
version 2
offset-list 0 out 10 FastEthernet0/1 ( add to the metric)
network 2.0.0.0
network 10.0.0.0
no auto-summary

After the offset change

*Mar  1 01:27:32.155: RIP: build update entries
*Mar  1 01:27:32.155:   1.0.0.0/8 via 0.0.0.0, metric 12, tag 0  - now we can see the metric change based on the offset list and the outgoing interface:
*Mar  1 01:27:32.159:   2.0.0.0/8 via 0.0.0.0, metric 11, tag 0
*Mar  1 01:27:32.159:   10.1.1.0/24 via 0.0.0.0, metric 11, tag 0
Note: We can also use the offset list with a ACL that matches a certain prefix to filter routes out. One issue with doing this is that if you specify a route ex: 192.168.1.0 /24. You will also be filtering out any other subnets in the prefix. For example – 192.168.1.0 /25 /26 / 27, etc …..