Outlines:

1-      Why we need multiple routing protocols? Which lead to need redistribution

2-      Redistribution meaning.

3-      Methods of Redistribution.

4-      Redistributing protocol  Vs Redistributed protocol

5-      Redistribution Issues.

6-      Redistribution command.

7-      Example of Redistribution.

 

 

Why we need multiple routing protocols and Redistribution?

 

A) Migration

-From FLSM to VLSM

-From fl–t design to hierarchical design (to facilitate route summarization which enhance network scalability)

B) Boundary between ASs

C) Different departments might require different routing needs

D) Unix host based routing (centre that contain UNIX servers) run RIP only, but your network require another protocol for inter routing

E) Mixed router vendor environment (use EIGRP on Cisco routers, use OSPF on non-Cisco router)

 

 Redistribution

 

•It is the mechanism that allow to connect different domains, so as the different Routing protocol can exchange and advertise routing updates as if they are a single protocol

•The redistribution is performed on the router that lies at the boundary between different domains or runs multiple protocols

 

Methods of Redistribution

 

•One way redistribution: redistribute networks learned by a certain protocol in a single direction

•Two way redistribution: redistributes all routes from a routing process to another and vice-versa

 

 

Redistributing protocol Vs Redistributed protocol

 

 

•Redistributing protocol: It is the native protocol that will transform another protocol to its form

•Redistributed Protocol: It is the non-native protocol that will be transformed to another protocol form

 

 

Redistribution issues

 

 

1) Administrative Distance: The redistributed protocol inherits the admin. Distance of the redistributing protocol
2) Metric: a seed metric (initial metric) is assigned to a redistributed route, then that metric is incremented according to the normal redistributing protocol policies
•The seed metric is adjusted from:

1-The default-metric command
2-The redistribution command using metric option or route map option (override the default-metric command)
Default Seed metrics

•If Redistributing is-RIP the metric is infinity
-If Redistributing is IGRP/EIGRP the metric is infinity
-If Redistributing is OSPF the metric is 20 (external type2) for all except 1 (type2).

So for RIP, IGRP and, EIGRP we need to set the metric manually using the metric commands.

 

Configuration of Redistribution

 

*Hereunder you can find the complete redistribution command:

•(config)#router <redistributing protocol>

 

•(config-router)#redistribute <redistributed protocol> [process id]
[metric <seed metric>] ! Initial metric for redistributed routes!

[match {internal/external 1/extenal 2}]

! If OSPF is redistributed: match O, OIA to be redistributed

or match OE1 to be redistributed

or match OE2 to be redistributed !

[Metric-type {1/2}] ! Metric type for redistributed routes into OSPF!
[Subnets] ! Consider subnets for redistribution into OSPF!
{Level-1/level-1-2/level-2}

! If IS-IS is redistributing:

redistribute routes to IS-IS as iL1 or iL2 (default is iL2)

If IS-IS is redistributed:

redistribute iL1 routes or iL2 routes or both !
[route-map <map name>] ! use route filter with redistribution !
[tag <tag name>]! Set tag for routes redistributed !

 

 

Redistribution Example

 

 

-Consider the following below example as we need the RIP subnets 11.11.11.0/24 be redistributed into OSPF and the OSPF subnets 44.44.44.0/24 be redistributed into RIP.

– Redistribution is performed only on R2 and R3 routers (boundary routers).

 

 

-I will focus only on the configuration of both R2 & R3 as the RIP & OSPF config. of R1 & R2 is very easy.

1- The default redistribution config. of R2 & R3:

R3 config:

 

 

 

 

R2 config:

 

 

NOTES:

-From the above config. we found that we need to set the metric when redistributing OSPF into RIP as we said before in the Redistribution Issue section.
– The command subnets under OSPF when redistributing the RIP is used to redistribute all OSPF subnets as without it only the class-full networks will be redistributed.

 

The Routing table of both R2 & R3:

 

 

 

Notes:

 

-After showing both RTG tables of R2 and R3 we found that R2 is learning the subnet 11.11.11.0/24 via next hop R1 as RIP route and the subnet 44.44.44.0/24 via next hop R4 as an OSPF route which is correct (Normal behavior).
-But if we focused on R3 RTG we will found that the subnet 44.44.44.0/24 is learned correctly via next hop router R4 as an OSPF route but the subnet 11.11.11.0/24 is wrongly learned via next hop router R4 as an OSPF instead of R1 as a RIP route due to the redistribution issue.
– We can solve this by many solutions I will choose the distribute-list one.

 

Solution:

-R2 config. after applying the distribution-list:

 

-R3 config. after applying the distribution-list:

 

Notes:

-First we configured two access-lists which permit only the subnets 11.11.11.0/24 and 44.44.44.0/24 (access-list 1&2) as shown above @ both R3 and R2 routers as the distribute-list needs to be active an access-list existence.
-From the config. shown we will find that we forced the RIP updates to permit only the subnet 44.44.44.0/24 to R1 router with the command (dist-list 2 out) and to accept only the subnet 11.11.11.0/24 from R1 with the command (dist-list 1 in), and forced the OSPF updates with the same behavior.
-Now we will show the RTG of both R3 & R2.

-R2 RTG after applying the distribution-list:

 

-R3 RTG after applying the distribution-list:

 

 

-We can found now that every  is normal which is  R2 is learning the subnet 11.11.11.0/24 via next hop R1 as RIP route and the subnet 44.44.44.0/24 via next hop R4 as an OSPF route.
-R3 is learning the subnet 11.11.11.0/24 via next hop R1 as RIP route and the subnet 44.44.44.0/24 via next hop R4 as an OSPF route.