Dynamic Routing—RIP

Table of contents

dynamic routing

Dynamic routing concept:

advantage:

shortcoming:

Division of agreements

Scope division

algorithm

RIP's neighbor

judging criteria

RIP algorithm

RIP protocol

Existing versions:

Difference between RIPV1 and RIPV2:

RIP packets

work process:

initialization

take over:

Judgment:                

RIP mechanism

RIP timer

RIP loop problem:

Solution to loop without problems:

RIP basic configuration:

RIP extension configuration:

1 RIP manual authentication ---- RIPV2

2. Manual summary of RIP

3. Silent interface

4. Accelerated convergence of RIP

5. RIP default route

dynamic routing

Dynamic routing concept:

        Devices running the same dynamic routing protocol, negotiate and finally calculate the route by communicating with each other through data packets

advantage:

1. Can reduce the amount of configuration

2. It can dynamically adapt to network changes and complete routing calculations by itself

3. Suitable for medium and large networks

shortcoming:

1. Calculated

Routing may not be what we want

Division of agreements

Scope division

AS: Autonomous system —> a kind of thinking of big and small

An AS runs a dynamic routing protocol

AS wants to communicate directly - BGP

algorithm

Link state routing protocol: OSPF (used by 80% of the world)

OSPF mainly transmits topology information

Distance vector routing protocols: RIP, EIGRP

EIGRP (Cisco-only): Pass-through routing table—passes on routing information in the routing table

RIP's neighbor

Neighbor of RIP : The device runs RIP, the device is adjacent, and can transmit RIP data packets through the command network segment

RIP actually only needs to pass two parameters: target network segment, cost: overhead

COST : The parameter used by the dynamic routing protocol for route selection

The cost of RIP only depends on the number of hops (the number of routers passed through)

judging criteria

The overhead between different routing protocols is meaningless (different criteria)

Comparison priority of different routes

The default priority of RIP is 100

The default priority of OSPF is 10

The default priority of static routing is 60. This set of standards is developed by Huawei

RIP algorithm

RIP Algorithm: Bellman Ford Algorithm

1. For R2, if I receive a piece of routing information that my local routing table does not have , I will directly refresh the routing information into my local routing table .

2. For R2, if I receive a piece of routing information that already exists in my local routing table , if the source is the same , I will refresh the routing information to my local routing table

3. For R2, if I receive a piece of routing information that already exists my local routing table , if the source is inconsistent , compare it according to the cost value carried in the passed routing information . If the cost value in the local routing table is small , it will not be refreshed . The reverse is also true.

The information COST transmitted by RIP == the cost value of the local routing table + 1

RIP protocol

Existing versions:

RIPV1

RIPV2 - IPV4 environment

RIPNG——IPV6

Difference between RIPV1 and RIPV2 :

1. V1 is a classful dynamic routing protocol, and V2 is a classless dynamic routing protocol

There is a category that does not carry the subnet mask in the process of transmitting data packets

No category is to carry the subnet mask in the process of transmitting data packets

2. V1 does not support manual authentication, V2 supports manual authentication (encryption)

3. V1 uses broadcast to send data packets, V2 uses multicast (address: 224.0.0.9 UDP port number: 520 ) to send data packets (multicast delivery is more flexible)

RIP packets

request: request package

response: response packet

work process:

initialization

                RIP will send a RIP request packet to each interface running the RIP protocol to request the routing table

receive :

                After the RIP neighbor receives the RIP request packet, it will encapsulate its own routing table (routing information) in the RIP response packet, and then send it to the peer through multicast 224.0.0.9

Judgment :                

                 According to the four situations stipulated by the Bellman-Ford algorithm, choose whether to add a table

RIP mechanism

RIP has a periodic update mechanism, so the response packet is also called an update packet ( updated every 30 seconds by default ). Therefore, when the network condition is good, the response packet is more than the request packet.

RIP is updated asynchronously : staggered peaks occur to avoid a large delay at a certain moment, occupying a lot of bandwidth, and better operation

RIP does not have a confirmation mechanism

RIP does not have a keep-alive mechanism

RIP timer

Periodic update timer: default 30 seconds

Dead timer: 180 seconds

Garbage collection timer: 120 seconds

When the 180-second expiration time expires, the router will first delete the network segment information from its own routing table, and at the same time start the garbage collection timer.

Send this network segment to its own neighbors, and at the same time, it will carry a cost of 16, which means that this network segment is unreachable.

It takes 180 seconds for RIP to completely delete a piece of routing information

RIP has a working radius : 15 hops (16 hops are considered unreachable)

RIP loop problem:

RIP loop problem:

                        asynchronous update

Solution to loop without problems:

                                16 jumps

trigger update

split horizon

        When I receive a route from an interface, it will no longer send the route information when updating

toxic reversal

        When the router receives a route from an interface, it will send the route information when updating, but it will carry the cost value and set COST to 16

(Huawei enables split horizon by default. If both mechanisms are enabled, it will be executed according to the logic of poisonous reversal)

RIP basic configuration:

[r1]rip ?

INTEGER<1-65535> Process ID Process ID, different process numbers are equivalent to different protocols

[r1]rip 1                                                    starts the RIP process

[r1-rip-1]version 2                                   select version

[r1-rip-1]network 1.0.0.0                         declaration ( only need to declare the network segment directly connected to itself )

RIP can only declare the main class

Publish route:

                Activate the interface—only the declared interface can send and receive RIP data packets normally

effect:

publish route

Activate the interface: only the declared interface can send and receive RIP data packets normally

RIP extension configuration:

1 RIP manual authentication ---- RIPV2

[r1-GigabitEthernet0/0/0]rip authentication-mode simple cipher123456- interface authentication

 simple: The password in the data packet sent by RIP is forwarded in plain text

 cipher: the local storage password is stored in plain text

Authentication algorithms must be consistent: Simple MD5 (two types)

[r2-GigabitEthernet0/0/0]rip authentication-mode md5 usual cipher 123456 Use MD5 to encrypt

2. Manual summary of RIP

[r1-GigabitEthernet0/0/0]rip summary-address 192.168.0.0 255.255.252.0

Configure the location where the RIP data packet is sent out , and the mask can only be written in dotted decimal, and the number cannot be written directly

3. Silent interface

This interface will only receive packets, but will not send RIP packets

[r1-rip-1] silent-interface GigabitEthernet 0/0/1 configuration location RIP process

4. Accelerated convergence of RIP

Change the RIP timer

[r1-rip-1]timers rip 10 60 40

10            60             40

Send cycle Life time Delete cycle

5. RIP default route

[r2-rip-1]default-route originate

Guess you like

Origin blog.csdn.net/2302_77035737/article/details/131636438