RIP PROTOCOL
In the smallest intranet manager can establish and modify routes manually.
The manager keep a list of networks and updates the forwarding tables when ever a new network is added to deleted from the network
RIP protocol remained in widespread use in early in the internet
Originally RIP was known by the software which implements it,routed.
RIP operation
The underlying RIP protocol is a straight forward implementation of distance vector routing for local networks.
RIP support two type of participants
A router running RIP active mode broadcast a updating messages every 30 seconds
Update message
It contain a set of pairs, each pair contain (IP network,integer distance to that network).
The update contains information taken from its forwarding infomation tables (FIB)
Hop Count metric
RIP use a hop count metric to measure distances, a router is defined to be one hop from directly connected network
one hop :- direct connection
two hop :- a network that is reachable through one another
cases
network 1→inter mediate network routes→network 5
If intermediates routes contian slower path with small hop number it is not taken
So selecting hop count minimum for shortest path routing is not produce optimal results
from above exmple although shrtest path is from through sattellite is not best to reach fastest way to to the other network
And some time two path can be exited with same hop count to the network .(1,1) means it can reach network 1 with distance 1.
(1,3) means it can reach network 1 with distance 3
In that case it take which advertise its routes first.
RIP use few rules to improve performance and reliability
Timer reset (180seconds)
Routes become invalid after 180 seconds if routes is not advertised within that time
So if a router fail it will get invalid path through that router within 180seconds
If a router failed and installed new router it will advertise its same routes for updating FIB and that time its timer is reset.
It uses hysteresis ,means that it will not replace routes with equal cost
Three kinds of error handled by RIP
Network 1 → R1→R2→R3
Network 1 X R1→R2→R3 ( network 1 failed )
Network 1 X R1⇌R2→R3 ( Here R2 is broadcasted its routing information with hop 2, this will taken by R1 router and add hop 3 and broadcast with distance 3 and this continues until the datagram's hop limit is reached )
Solution is like good news travel quickly and bad news travel slowly
In the smallest intranet manager can establish and modify routes manually.
The manager keep a list of networks and updates the forwarding tables when ever a new network is added to deleted from the network
RIP protocol remained in widespread use in early in the internet
Originally RIP was known by the software which implements it,routed.
RIP operation
The underlying RIP protocol is a straight forward implementation of distance vector routing for local networks.
RIP support two type of participants
- ACTIVE :-Active partipants advertise thier routes to others
- PASSIVE :- Passive participants listen to to RIP messages and updates their forwarding tables
A router running RIP active mode broadcast a updating messages every 30 seconds
Update message
It contain a set of pairs, each pair contain (IP network,integer distance to that network).
The update contains information taken from its forwarding infomation tables (FIB)
Hop Count metric
RIP use a hop count metric to measure distances, a router is defined to be one hop from directly connected network
one hop :- direct connection
two hop :- a network that is reachable through one another
cases
network 1→inter mediate network routes→network 5
If intermediates routes contian slower path with small hop number it is not taken
So selecting hop count minimum for shortest path routing is not produce optimal results
from above exmple although shrtest path is from through sattellite is not best to reach fastest way to to the other network
And some time two path can be exited with same hop count to the network .(1,1) means it can reach network 1 with distance 1.
(1,3) means it can reach network 1 with distance 3
In that case it take which advertise its routes first.
RIP use few rules to improve performance and reliability
Timer reset (180seconds)
Routes become invalid after 180 seconds if routes is not advertised within that time
So if a router fail it will get invalid path through that router within 180seconds
If a router failed and installed new router it will advertise its same routes for updating FIB and that time its timer is reset.
It uses hysteresis ,means that it will not replace routes with equal cost
Three kinds of error handled by RIP
- Algorithm does not cleaarly detect forwardind loops,so RIP must either trust or take precautionto prevent such loops
- To prevent instabilities RIP must use a low value for maximum possible distance (16),manger must divide the intranet into sections or use an alternatice protocol
- The distance vector alogorithm can create slow convergence or count o infinity in which inconsistance occur , because routing propagation occur slowly across network , choosing small infinity(16) helps to limit convergence but dot not eliminate it.
Network 1 → R1→R2→R3
Network 1 X R1→R2→R3 ( network 1 failed )
Network 1 X R1⇌R2→R3 ( Here R2 is broadcasted its routing information with hop 2, this will taken by R1 router and add hop 3 and broadcast with distance 3 and this continues until the datagram's hop limit is reached )
Solution is like good news travel quickly and bad news travel slowly
- Split horizon update,It does not propagate information about a router back over the same technique
- Hold down:- wait long enough to ensure that all machine receives the bad news and not mistakenly accept a message that is out of date,Hold down set time for RIP is set to 60seconds, i.e twice as long a normal update message
- Poison reverse :- Once a connection disappear the router advetising the connection retain the entry for several update period and include an infinit cost route in its broadcast
- Triggered update :-Triggered update mechanism forces a router to broadcast the changes in routing information immediatly after receiving bad news
Q.I.c.Poison reverse combined with triggered updates can solve slow convergence problem created byRIP. Justify the given statement.(5 mark)[B. Tech. Degree VII Semester Examination November 2012 CS/IT 703 ADVANCED COMPUTER NETWORKS (2006 Scheme)]
No comments:
Post a Comment