[BGP] small scale experiment

Catalog

1. The topology ：

2. The experimental requirements ：

（1） Network wide accessibility

（2） Use BGP 

3. Experimental thinking ：

（1） To configure IP Address

（2） stay AS2 Zhongqi RIP（ dynamic ）

（3）R1,R2 Write static    R3,R4 Write static

（4） rise BGP

（5） test

4. Configuration command ：

To configure IP：

Static configuration ： （ With R4,R5 For example ）

RIP To configure （ Dynamic protocols are used for AS2 Inside ） 

BGP To configure ：

test ：

R1 Loopback ping R4 Loopback ：

R1 Loopback ping R5  Loopback ：

Summary ：

Add ：

---

1. The topology ：

2. The experimental requirements ：

（1） Network wide accessibility

（2） Use BGP 

3. Experimental thinking ：

（1） To configure IP Address

（2） stay AS2 Zhongqi RIP（ dynamic ）

（3）R1,R2 Write static    R3,R4 Write static

（4） rise BGP

（5） test

4. Configuration command ：

To configure IP：

 

 

 

 

Static configuration ： （ With R4,R5 For example ）

 

RIP To configure （ Dynamic protocols are used for AS2 Inside ） 

 

 

BGP To configure ：

 

 

 

 

 

test ：

R1 Loopback ping R4 Loopback ：

 

R1 Loopback ping R5  Loopback ：

Summary ：

  This small experiment ： Use of the BGP The basic configuration of , The follow-up editorial meeting will release relevant BGP Detailed knowledge of .

Add ：

1. BGP: Border gateway routing protocol    Classless path vector EGP agreement ;

2. EGP Class of routing protocols , be used for AS And AS Share routing entries between ;

3. AS Number 0-65535 Standard No , among 1-64511 public     64512-65535 private

4.        32 Bit binaries are built as extension numbers ;                                                               

5. http://www.cidr-report.org  IANA

6. Design features ：

7. 1、 reliability — A large number of routes are shared , Cancel periodic update       TCP 179 Port operation

8.    TCP Unicast communication only      BGP The protocol requires unicast to establish a neighbor relationship （ Manually specify the neighbor's ip Address ）

9.    BGP You can build a neighbor through indirect connection （peer）,BGP Carried on IGP above ;

10. 2、 Controllability --- Jianlin 、 Route announcement 、 Routing control （ Interference in routing ）-- It is extremely easy to

11. 3、AS-BY-AS    With a AS For a jump

12. BGP: Border gateway routing protocol  

13. One 、 Working in AS And AS Between EGP Category agreement ;

14. AS— Autonomous systems    standard AS The length of No. is 16 position —0-65535 

15.   among 1-64511 public     64512 -65535 private

16.     Expand AS Number length 32 position    about 42 Billion

17. CIDR Report

18. BGP The agreement is different AS Shared routes between , Isolate two when working AS The internal network of , Form an independent interaction space ; At the same time, powerful strategies can be implemented to interfere with the route selection ;

19. IGP Pursuit ： Fast convergence       Choose a path （ acyclic ）     Less occupied resources   hop-by-hop

20. EGP Pursuit ： Controllability         reliability          AS-BY-AS

21. Controllability —BGP The protocol is not the generator of the route , For the carrier only , And AS There should be a large number of links to ensure stability ; so BGP It is difficult for the routing items passed by the protocol to default to the best routing ;-- The administrator needs to carry out a large number of policies on routing to interfere with routing ;

22. reliability —BGP Like the distance vector protocol, the protocol only passes routing entries , But the amount of updates is still very large ;

23.         In order to avoid the occupation of equipment resources, incremental update is selected --- Only trigger 、 No period   

24.         Therefore, it is necessary to ensure that all routes are delivered in place — reliable     therefore BGP Based on the agreement TCP Work

25.         TCP The triple handshake condition of makes TCP Can only work based on unicast , meanwhile BGP Non direct connection and neighbor building are required ;

26.         BGP be based on IGP above , First IGP send BGP Equipment room IP Can be up to , then BGP Can be based on TCP Establish indirect neighbors ;

27. AS-BY-AS— With a AS For a jump , Will a AS As a whole ;

28. Two 、BGP characteristic ：

29. Classless path vector ----- Upgraded version of distance vector ---AS--BY--AS

30. Use unicast updates to send all information ; be based on TCP 179 Port operation

31. Incremental updating -- Trigger only aperiodic

32. Have rich attributes to replace IGP Route selection based on measurement ---- Multiple parameter control protocol

33. You can implement a strong strategy for traffic in and out -- Controllability

34. The default is not used for load balancing ----- Only one optimal path is generated through various routing rules

35. BGP Support authentication and aggregation （ Summary ）

36. 3、 ... and 、BGP Data packets     be based on TCP Of 179 Port operation ; so BGP All data packets in the protocol need to be in tcp After the session is established , be based on TCP To ensure the transmission and reliability ;

37. First, through TCP Three handshakes to find a neighbor ;

38. Open       Only responsible for the establishment of neighborhood relations , Normal receiving and sending can be done once ; carry route-id,hold time Default 3min

39. Keeplive    Keep alive     cycle 1min Query whether the neighbor relationship exists ; Actually keep alive TCP conversation ;

40. Update     Carry routing entries      Target network number + Various attributes

41. Notification   Send and receive error data ;

42. Four 、BGP Working process of

43. The administrator defines the neighbors ip Address , The premise is that the neighbors ip Can be up to ; --- IGP Can be up to

44. start-up BGP After the agreement , First, the target port is 179 Conduct TCP Three handshakes to establish TCP Conversation ;

45. After the session is established , Send and receive open Message to establish neighbor relationship , Generate neighbor table ;

46. After the neighbor relationship is established , Use between neighbors update Shared routing entries , After receiving and sending the routing information , Locally generated BGP surface ;

47. Bgp The table loads all routing entries sent and received locally ;

48. Then the router will BGP The best path in the table （ Not necessarily the best choice , Only for BGP The parameters are optimal ） Loaded in the routing table ;

49. Convergence complete ! only keeplive Keep the cycle alive ;

50. Structural mutation ：

51. New network segment – BGP The update source device will use update To inform all the local neighbors

52. Disconnect the network segment – BGP The update source device will use update To inform all the local neighbors

53. Unable to communicate – 3min hold time When the time comes, disconnect the neighbors and TCP conversation , Then delete all the information learned from the neighbor ;

54. BGP Updating the source device does not necessarily directly connect to the disconnected or newly added network segment devices , Instead, the first one will publish the route to BGP Router in the protocol ;

55. 5、 ... and 、 Professional terms ：

56. EBGP External BGP Neighborhood , The two are in different positions AS Of BGP The neighbor relationship established between routers

57. IBGP  Inside BGP Neighborhood , Two are in the same AS Internal BGP The established neighbor relationship between routers

• It's not easy to create , Thank you for your support , I hope it can help everyone who works hard ！！