MGRE GRE OSPF process in hcip

 

 

One 、 Introduce
Physics line ：

What happened ：

1, cost ;

2, Geographical constraints

VPN（ Virtual private network ）：

VPN The core of Technology ： Tunnel technology

The core of Tunnel Technology ： Packaging technology

Two 、GRE： General routing encapsulation technology

The way we want to go ：

The real way ：

Tunnel technology ： At both ends of the tunnel, a data channel is established in the public network through encapsulation and de encapsulation , Use this data channel to enter                       Line transmission .

GRE Configuration method :

1, Create a tunnel interface

      [r1] interface Tunnel 0/0/0

      [r1-Tunnel0/0/0]

2, Interface configuration IP Address

      [r1-Tunnel0/0/0] ip address 192.168.3.1 24

3, Define encapsulation method

      [r1-Tunnel0/0/0] tunnel-protocol gre

4, Define the content of the package

      [r1-Tunnel0/0/0] source 12.0.0.1

      [r1-Tunnel0/0/0] destination 23.0.0.2

3、 ... and 、NHRP： Next - Hop Resolution Protocol
NHRP Server for ： NHS, Next hop to the resolution server -

NHRP principle ： You need to select a device with a fixed exit physical address in the private network as NHS, All remaining branches should                         Know the tunnel address and physical address of the center , then ,NHRP All branches are required to have their own physical interfaces and tunnels                       Interface IP The mapping relationship of the address is sent to NHS, If the physical address changes , You need to resend . this                       sample NHS You can get the mapping relationship of the addresses of all branches . If the branches need to communicate with each other , You need to                       Apply for the mapping table like the center --- This architecture is called hub-spoke framework .

Four 、MGRE Configuration of ：
Configuration of the center :

1, Create a tunnel interface

     [r1] interface Tunnel 0/0/0

2, Interface configuration IP Address

     [r1-Tunnel0/0/0] ip address 192.168.3.1 24

3, Define encapsulation method

     [r1-Tunnel0/0/0] tunnel-protocol gre p2mp

4, Define the content of the package

     [r1-Tunnel0/0/0] source 15.0.0.1

5, establish NHRP Domain

     [r1-Tunnel0/0/0] nhrp network-id 100

Configuration of branches :

1, Create a tunnel interface

      [r1] interface Tunnel 0/0/0

2, Interface configuration IP Address

      [r1-Tunnel0/0/0] ip address 192.168.3.1 24

3, Define encapsulation method

      [r1-Tunnel0/0/0] tunnel-protocol gre p2mp

4, Define the content of the package

     [r2-Tunnel0/0/0] source GigabitEthernet 0/0/1

5, Join the center to create NHRP domain

     [r2-Tunnel0/0/0] nhrp network-id 100

6, Report information to the center

     [r2-Tunnel0/0/0] nhrp entry 192.168.5.1 15.0.0.1 register

Tunnel address   Physical interface address

notes ：[r1-Tunnel0/0/0] display nhrp peer all       -- see NHRP Registration of neighbors

notes ：MGRP Environment when sending data , It is still a point-to-point tunnel , So the data transmission is still point-to-point           transport . therefore ,MGRE The environment is similar to NBMA Environment .

5、 ... and 、RIP Realize in MGRE Environmental problems :
1, Only the center gets the routing information of the branch , The branch did not get

reason ：rip It is to transmit routing information in the form of broadcast and multicast ; and MGRE Unicast transmission

resolvent ： Turn on the pseudo broadcast on the center

                  [r1-Tunnel0/0/0] nhrp entry multicast dynamic

2, After the center turns on the pseudo broadcast , The branch can only receive the routing information of the center , But there are no branches .

reason ：RIP The horizontal segmentation mechanism of

resolvent ： close RIP The horizontal division of

                  [r1-Tunnel0/0/0] undo rip split-horizon

6、 ... and 、OSPF（ Open shortest path first protocol ）
RIP Version of ： RIPV1,RIPV2（IPV4）,RIPNG（IPV6）

OSPF Version of ：ospfv1( Days off in the Laboratory ),ospfv2（IPV4）,ospfv3（ IPV6）

RIPV2 and OSPFV2 Similarities ：

1,RIPV2(224.0.0.9) and ospfv2(224.0.0.5,224.0.0.6) They all send information in the form of multicast .

224.0.0.X Format multicast address ： Local link multicast , And TTL Set to 1, It can only communicate within the range of direct connection

2,RIPV2 and OSPFV2 Are called classless routing protocols .

No category ： The subnet mask will be carried when sending routing information

notes ：OSPF What is passed is topology information

OSPFV2 It is called classless Routing Protocol ：

Because the behavior of routing exists between domains

3,OSPFV2 and RIPV2 All support manual authentication

4,OSPFV2 and RIPV2 All support equal overhead load balancing

RIP and OSPF Difference point :

RIP It can only be used in small networks ,OSPF It can be applied to medium and large networks

RIP The reason why it can only be used in small networks ： It takes up a lot of resources

OSPF It can be applied to medium and large networks ： because OSPF Support structured deployment

Structured deployment ： Is the division of regions , Regional division is to divide large areas into small areas

The purpose of regional division ： Transfer topology information within the region , Transfer routing information between regions .

Transfer routing information between regions ：

notes ：OSPF If the network has only one area , Such a network is called a single area OSPF The Internet ; If there are multiple areas , said           For multiple areas OSPF The Internet .

Area border router （ABR equipment ）： Belong to multiple areas at the same time , An interface corresponds to an area , And there is an interface in                                                       Area 0 in .

notes ： There can be multiple... Between regions ABR equipment , One ABR It can also correspond to multiple regions .

OSPF Requirements for area division ：

1, There must be... Between areas ABR equipment

2, Region division must be based on star topology , The middle region of star topology is called backbone region .

Area ID(area ID)： Distinguish and calibrate ospf Different areas in the network , from 32 Bit binary structure

Area ID(area ID) The way of expression ：

1, Dotted decimal means ;

2, Use decimal system directly to express , The backbone area ID Defined as area 0（ necessary ）.

7、 ... and 、OSPF working process ：
1,OSPF Data packets of （ Five kinds of bags ）

hello package ： Discover in periodic form , The role of building and preserving neighborhood relations

hello Time ：10S ( or 30S)              

notes ： Under different network environments ,hello Time is different , Ethernet （BMA） stay BMA The default under the network hello Time is         10s, But in other networks ,hello Time is 30s

Time of death （Dead time）：4 Times hello Time

OSPF And RIP difference ：

RID（RouteID）（ Used to mark the router ） characteristic ：

1, The whole network is unique （OSPF In the network ）;

2, use the same pattern ： The requirements must be in accordance with IP Address format to design , by 32 Bit binary structure .

RID Access method ：

1, Manual configuration ： Just meet the above two conditions

2, Automatic generation

How to generate automatically ：

1, First, check whether the device is equipped with a loopback interface , If it exists, select the loopback interface IP Address as RID; If there are multiple loop back connections         mouth , Then the maximum value will be selected （ From front to back ） As RID.

2, If there is no loopback interface , Then the... Of the physical interface of the device will be taken IP Address as RID, If there are multiple physical interfaces , be           The one with the largest value will be selected as RID.

DBD package ： Database description message

database ：LSDB( Link state database ), For storage LSA, Link status notification

LSR package ： Link status request message , according to DBD Package comparison , Based on local unknown LSA Message sending request .

LSU package ： Link state update message , Really carry LSA Packets of information

LSAck package ： Link status acknowledgement message , To ensure the reliability of the transmission

notes ：OSPF There is 30min A periodic update .

2,OSPF State machine

 

TWO-WAY： It marks the establishment of neighbor relations

( Matching conditions )： The match is successful , Then you can enter the next state , If you fail , Will stay in the neighborhood , Use only                            Hello The package is kept alive periodically .

 

Principal subordinate relationship election ： By comparison RID To carry out ,RID Big for the main , The master can enter the next state first .

notes ： Use here DBD Package to complete the master-slave relationship election , Mainly to distinguish from the previous neighbor state .

 

FULL state ： It marks the establishment of adjacency . The purpose is to distinguish from the neighbor state .

notes ： Neighbor status can only be used hello The package is kept alive periodically , And the adjacent state can send and receive LSA Information .

down state ： start-up ospf after , issue hello After the package, go to the next state init( initialization ) state ： received Hello In bag                         Include local RID, Then enter the next state            

Two-way( Two-way communication ) state ： It marks the establishment of neighbor relations

( Entry conditions match )： Matching failure , Will stay in the neighborhood , Use only Hello The package is kept alive periodically ; If the match is successful, enter                               Next state

exstart( Pre start ) state ： Use a that does not carry data DBD Package for master-slave relationship election ,RID Big for the main , Priority can be given to                                       Go to the next state

exchange( Quasi exchange ) state ： Use... With directory information DBD Package for directory sharing

loading( load ) state ： Based on DBD package , Use LSR/LSU/LSACK Three kinds of packet acquisition are unknown LSA                                      （ Link status notification ） Information .

FULL state ： It marks the establishment of adjacency .

3,OSPF Working process of

After the startup configuration is completed ,ospf Multicast will be provided to all interfaces running the protocol locally .

224.0.0.5 In the form of hello package ;hello The bag will carry your own local RID And local known neighbors RID; after , Record the collected neighborhood relationships in a local table （ Neighborhood watch ）.

After the neighbor relationship is established , Condition matching will be performed . Failure , Stay in the neighborhood , Use only hello The package is kept alive periodically ;

If the matching is successful, the adjacency relationship will be established . First , Use a that does not carry data DBD Package for master-slave relationship election , after , Use... With information DBD Package shared database directory information . after , Local use LSR/LSU/LSACK Three kinds of packets get unknown LSA Information . after , Complete the establishment of local database , Generate database tables （LSDB）.

Last , Based on LSA Information , Generate directed graph and shortest path tree , after , Calculate the local routing information to the unknown network segment . Add these routing information to the routing table .

After convergence ,ospf Still every 10s(30s) send out hello The package is kept alive periodically ; every other 30MIN Make a periodic update .

8、 ... and 、OSPF Structural mutation ：
1, Add a new network segment ： Trigger update , Change the information immediately through LSU package Pass out , need ACK confirm .

2, Disconnect a network segment ： Trigger update , Change the information immediately through LSU package Pass out , need ACK confirm .

3, Unable to communicate ：dead time

4,OSPF The basic configuration of

1, start-up ospf process

     [r1] ospf 1 router-id 1.1.1.1    ---1 Bit process number

     [r1-ospf-1]

2, Create area

    [r1-ospf-1] area 0

    [r1-ospf-1-area-0.0.0.0]

3, Declare

1, Activate interface

2, Publish route

[r1-ospf-1-area-0.0.0.0] network 12.0.0.1  0.0.0.0            --0.0.0.0 Is unmasked

Unmask ： Continuous 0 And continuous 1 form ,0 The corresponding bit is immutable ,1 Corresponding to variable , You can make accurate announcements , It's fine too                   Declare the scope

[r1] display ospf peer                                    --- see ospf My neighbor table

[r1] display ospf peer brief                            --- View neighbor profile

[r1] display ospf lsdb                                     --- View database tables

[r1] display ospf lsdb router 2.2.2.2              --- Expand a LSA Information

Huawei equipment definition ospf The default priority of the protocol is ：10

COST= Reference bandwidth / Real bandwidth

The default reference bandwidth of Huawei equipment is 100Mbps

notes ： If the cost value is less than 1 Decimals of , Then directly follow 1 To calculate ; If it is greater than 1 Decimals of , Then take the integer part directly .

Modify the reference bandwidth ：

[r1-ospf-1] bandwidth-reference 1000

notes ： Modifying the reference bandwidth requires that all OSPF The devices in the network are all changed to the same .