RIP

RIPv1
RIPv1 is a classfull routing protocol. CLASSFULL ROUTING DOESN'T INCLUDE THE SUBNET MASK INFORMATION IN ITS ROUTING INFORMATION UPDATES. because RIP is a classful routing. when you enter a subnetted network, the IOS will automatically convert it to the corresponding classful network. for example, network 192.168.1.32 will be converted to 192.168.1.0. Because of this behaviour, there is a design issue in RIP. there shouldn't be two subnetted networks with the same classful network address reached by different interface (discontiguous subnet) from a router. RIP advertise classful network. the routing table entry in this router will be unstable (if the cost of both subnets are the same) or this router will send the packets load balanced through each route (because there are two entries for the same major network).

RIPv1 broadcast its routing entires through update packets. a single update packet, can hold up to 25 routing entries.

RIP implements split horizon, so routes learned from an interface is not sent when sending updates to that interface.

you can stop a router running a routing protocol (the command discussed next is supported by all routing protocol), in this case RIP, from sending routing advertisement such as periodic updates to an interface by entering the following command at router's global configuration level.

passive-interface interface-type interface-number

interface-type : type of the interface (FastEthernet, Serial, ..).
interface-number: number of the interface.

you might want to enter this in a case where there's no neighbor router running the same routing protocol on that interface.

Routers running RIPv1 are limited to using the same subnet mask for all subnets with the same classful network.

************
RIP Timers
************
RIP use periodic updates to update its routing table. therefore there are some timers associated with RIP. Here are the timers

Periodic Updates - 30 secs
Invalid Timer - 180 secs. if a route is not refreshed up to 180secs. the route is marked invalid by making it unreachable (rip do this by changing the metric to 16)
Flush Timer - 240 secs. 60 secs longer than invalid time. after flush timer expires, the route is removed from the routing table.
Holddown Timer - 180 secs. if a router learns about a route from another router that it's unreachable, the route still stay in the routing table up to the holddown timer before it's removed to give time for every router to learn about the unreachable route. any advertisement with the same or worse metric reporting the route will be ignored. however, if an advertisement telling a better metric, the entry will be renewed and the holddown timer will be cleared.

RIPv1 sends updates to 255.255.255.255, which means a local broadcast address. any router will block packets destined to this destination from going out.

however RIP might recognize a subnet network. examine the following routing table.

172.30.0.0/24 is subnetted, 3 subnets
R 172.30.1.0 [120/1] via 172.30.2.1, 00:00:03, Serial0/0/0
C 172.30.2.0 is directly connected, Serial0/0/0
C 172.30.3.0 is directly connected, FastEthernet0/0

You can see that the route to 172.30.1.0 is gotten from R (RIP). How can RIP know that this network's subnet mask is 24? this can happen if the network advertised and the interface that received the advertisement have the same major network (classful network). in this case, the interface receiving the 172.30.1.0 advertisement has an ip address with network id of 172.30.x.x

RIP may have more than one route for a destination, if there's more than one route with equal cost to the same destination. RIP will use one of these for load balancing. if you enter "show ip protocols" command. you can see this information in the RIP section.

Maximum path: 4

this means that there can be up to 4 redundant equal cost path to the same destination.


Boundary Router : any router with interfaces in more than one major classful network.
boundary router may want to summarize the subnets of one major network to another major network (in fact that RIP only advertise major networks)

if the route entry for an update sent out to a different major network interface. then the network addres in the route entry is summarized to the classful network and then advertised. so the in the receiving router, the entry in the routing table for those subnets is only one, that is the major network of the subnets.

when an RIP interface first come up, it will sent a request message to neighboor routers. neigboor routers running RIP will send response message to the requesting router. it evaluates each route entry. If a route entry is new, the receiving router installs the route in the routing table. If the route is already in the table, the existing entry is replaced if the new entry has a better hop count. The startup router then sends a triggered update out all RIP-enabled interfaces containing its own routing table so that RIP neighbors can be informed of any new routes.

RIP doesn't support VLSM either. in a case where there are subnets from the same major network is applied with different subnets. only the subnets with the same subnet mask as the interface address of the outgoing update are advertised.

Supernet route will not be advertised by classful routing protocol including RIP. imagine in a case that you're creating a static default route with the destination address is a supernet. and you want RIP to redistribute this default route through the "redistribute static" command. however, because it's a supernet route, that default route won't be advertised to neighboring routers.

if you want to propagate static default route through RIP, you can enter the following command at the router configuration level

default-information originate



RIPv2
classles routing protocol

A ROUTER RUNNING RIPv1 CAN STILL PROCESS AN RIPv2 MESSAGE, IT WILL JUST IGNORE RIPv2 SPECIFIC FIELD.

RIPv2 Message (packet) contains subnet information, next-hop (The Next Hop address is used to identify a better next-hop address).


0 comments:

top