Article Directory
1. Packet forwarding
In the so-called packet forwarding, the data is transmitted in the form of packets. If the source end wants to send a large amount of data, it will be divided into several packets. Each packet contains some control information in addition to user data, so that the packet can be correctly transmitted to the destination end. Packet transmission adopts the store-and-forward method, that is, the network node sends the packet to the next node according to the packet control information. After the next node receives the packet, it temporarily saves it and waits for transmission in a queue, and then transmits the packet to the next node according to the packet control information until it reaches the final destination.
In the Internet, routers use the packet forwarding mechanism to realize the physical transmission of IP packets and datagram forwarding. As shown in the figure below, packet forwarding can be divided into direct forwarding and indirect forwarding according to whether the source address (IP address of the network layer) and destination address (IP address of the network layer) of the packet are in the same network.
When the source address and destination address of the packet are in the same network, the source host can directly forward the packet to the destination host, or when the destination router forwards the packet to the destination host, it is also direct forwarding. Otherwise, when the source address and destination address of the packet are not in the same network, it needs to be forwarded through routing lookup, that is, indirect forwarding. In the case of indirect forwarding, the router needs to find the next hop address to be forwarded from the routing table, and then process the IP packet and forward it to the next hop until it reaches the destination host.
Schematic diagram of packet forwarding
2. Routing
1. The basic characteristics of an ideal routing algorithm
An ideal routing algorithm needs to meet some basic characteristics: simplicity, correctness, stability, robustness, fairness, optimality, and efficiency.
Simplicity means that the routing algorithm cannot be too complex.
Correctness means that the routing algorithm must first ensure its own correctness, and at the same time, it should meet the requirements of data transmission, so as to meet the needs of network users' business work.
Stability means that the routing algorithm should respond appropriately to changes in the network state, neither too sensitive nor too slow.
Robustness means that the routing algorithm can adapt to changes in the network state such as topology and traffic, keep the network running as normal as possible, and continue the interrupted data transmission without restarting the network or terminating the host data transmission.
Fairness means that the routing algorithm should be equal to all users (except for a few users with high priority).
Optimality means that the routing algorithm should be able to find the best route, so that the average delay of the packet is the smallest and the throughput of the network is the largest.
Efficiency means that due to the needs of the routing algorithm, each node in the network will bear a certain amount of processing overhead, and at the same time generate some transmission overhead, which will affect the efficiency of the network to a certain extent.
2. Metrics for Routing Algorithms
- Bandwidth: refers to the transmission rate of the link (for example, the bandwidth of a T1 link is 1.544Mbit/s)
- Load: Refers to the amount of traffic passing through network resources (such as routers or links) per unit time.
- Delay: refers to the time it takes for a packet to reach the destination node from the source node. Latency depends on factors such as the bandwidth of intermediate links, the port queues of each router passing through, network congestion, and physical distance.
- Hop count: refers to the number of routers that a packet passes through from the source host to the destination host. A path with fewer hops is better. Reliability: refers to the bit error rate of each network link during transmission. -
- Cost (overhead): It can be an arbitrary value, generally based on cost, bandwidth overhead or other metrics, and can be specified by the network administrator.
3. Routing algorithm classification:
(1) Static routing algorithm.
Static routing has the characteristics of simplicity and low overhead, and is suitable for simple and small-scale networks, because in this network environment, it is easier for network managers to grasp the topology of the network, and the workload of manually configuring routing information is not large. It is also beneficial to ensure the security and confidentiality of the network, because static routing is manually configured, so there is no problem that routing information is intercepted and analyzed due to the exchange of their own routing tables between routers, thereby leaking information such as network topology and network addresses.
(2) Dynamic routing algorithm.
The dynamic routing algorithm is also called adaptive routing. Its routing information is obtained by exchanging routing information between connected routers, and then calculated and optimized according to a certain algorithm. Its characteristic is that it can better adapt to the change of network status, but it is more complicated to implement and the cost is relatively large.
4. IP routing selection and routing aggregation (emphasis)
The network in the figure includes access, aggregation, and core layers, which are connected by access layer routers, aggregation layer routers, and core layer routers respectively. There are two main types of connection links between routers, namely serial lines (marked with S1, S2) and Ethernet lines (marked with F0, F1, F2). The core layer router R1 is connected to the aggregation layer routers R2 and R3 respectively through two serial lines S1 and S2: the router R2 is connected to the access layer routers R4 and R5 through the Ethernet interface; the router R3 is connected to the access layer routers R6 and R7 through the Ethernet interface. Access layer routers R4, R5, R6, and R7 are respectively connected to 8 subnets of 200.200.0.0/24-200.200.3.0/24, 200.200.128.0/24~200.200.131.0/24. Then in this network, plus the two subnets connecting the core router and the aggregation router, there are 12 subnets in total. Correspondingly, there should be 12 routing entries in the routing table of core router R1, as shown in the table.
Schematic diagram of route aggregation
In this network, if the static routing table is used, it is necessary to manually add routing entries one by one in the 7 routers in the figure, so the total number of routing entries that need to be input is 12×7=84. It can be seen that the workload is relatively large, and when any router or link changes, the routing tables of the seven routers have to be reconfigured, so only dynamic routing protocols can be used to automatically establish and update the routing tables.
Routing table of the core router
| routing | interface | routing | interface |
|---|---|---|---|
| 200. 200,.200.0/30 | S1 | 200.200.2.0/24 | S1 |
| 200. 200. 201.0/30 | S2 | 200.200,3.0/24 | S1 |
| 200. 200.100.0/28 | S1 | 200, 200. 128.0/24 | S2 |
| 200. 200. 101.0/28 | S2 | 200. 200. 129.0/24 | S2 |
| 200.200.0.0/24 | S1 | 200. 200. 130./24 | S2 |
| 200.200.0.0/24 | S1 | 200,.200.131.0/24 | S2 |
| 200. 200.200.0/30 | S1 | 200.200.2.0/24 | S1 |
| 200. 200. 201.0/30 | S2 | 200.200,3.0/24 | S1 |
| 200. 200.100.0/28 | S1 | 200, 200. 128.0/24 | S2 |
| 200. 200. 101.0/28 | S2 | 200. 200. 129.0/24 | S2 |
| 200.200.0.0/24 | S1 | 200. 200. 130./24 | S2 |
| 200.200.0.0/24 | S1 | 200,.200.131.0/24 | S2 |
We observed that the routing table entries of the routing table of the core router can aggregate 12 routing table entries to a certain extent according to the "longest prefix matching" method, especially the 8 class C network routing entries connected to the access layer .
First of all, for 200.200.0.0-200.200.3.0, the first 16 digits of the two are exactly the same, so the summary route must also contain the word "200.200.XX", and then we observe their third field: 0=00000000 1= 00000001 2= 00000010 3=00000011 It can be seen that they have completely The same first 6 bits, plus the same first 16 bits before, a total of 8+8+6-22 bits are exactly the same, and these same bits constitute the network bits
of
the
summary
address
. There are 32-22-10 bits left, which are the host bits of the route after summarization. When expressing the network number of the summary route, the host bits are all represented by 0. Therefore, the aggregated route obtained by 200.200.0.0-200.200.3.0 through route aggregation is 200.200.0.0/22. In this way, the routing table is reduced from the original 12 routing entries to 6, and the simplified routing table is as shown. This is the whole process of route aggregation.
The routing table of the aggregated core router
| routing | interface | routing | interface |
|---|---|---|---|
| 200. 200,.200.0/30 | S1 | 200.200.101.0/28 | S2 |
| 200. 200. 201.0/30 | S2 | 200.200,0.0/22 | S1 |
| 200. 200.100.0/28 | S1 | 200, 200. 128.0/24 | S2 |
| 200. 200. 200.0/30 | S1 | 200. 200. 101.0/28 | S2 |
| 200.200.201.0/30 | S2 | 200. 200. 0.0/22 | S1 |
| 200.200.100.0/28 | S1 | 200,.200.128.0/22 | S2 |
Assuming that R1 receives a data packet with a destination address of 200.200.3.129, R1 needs to find an optimal path in the routing table, and follow the route lookup steps to match the destination address 200.200.3.129 with the routing entries in the routing table: the binary table of 200.200.3.129/32 is 11001000.11001000.00000011.1000 0001, compared with the binary representation of 200.200.0.0/22 11001000.11001000.00000000.00000000, it can be seen that the destination address 200.200.3.129 and 200.200.0.0/22 have exactly the same 22-bit address prefix, so the core router R1 will find the exit according to its corresponding routing entry, and finally send the packet Forward it from S1 port.
5. Real questions on router technology over the years
Knowledge points related to router technology
1. The packet forwarding capability of a router is related to the number of ports, port speed, packet length and packet type. (no port type)
2. High-performance routers generally adopt a switchable structure, and traditional core routers use a common-word backplane structure.
3. The packet loss rate is one of the performance indicators to measure the router when it is overloaded. ("routing table capacity" is not)
4. Throughput refers to the packet forwarding capability of the router, including port throughput and overall machine throughput. Backplane capacity determines router throughput. (It is not the throughput that determines the backplane capability of the router)
5. The burst processing capability is measured by the maximum sending rate at which data packets are sent at the minimum frame interval without loss, not just the minimum frame interval value.
6. Voice and video services have high requirements on delay and jitter.
7. The quality of service of the router is mainly manifested in the queue management mechanism, port hardware queue management and supported QoS protocol types. (not packet forwarding efficiency)
8. The router determines the packet forwarding path through the routing table.
9. The router's queue management mechanism refers to the router's queue scheduling algorithm and congestion management mechanism.
Relevant real questions over the years
Question 1: Among the following descriptions about router technical indicators, the correct one is ().
A. The packet loss rate is usually a measure of the performance of a router when it is overloaded
B. The throughput determines the backplane capability of the router
C. Voice and video services do not have high requirements for delay and jitter
D. Burst processing capability is measured by the minimum frame interval value
Correct answer: A
Reference analysis:
[Analysis] The performance indicators of routers mainly include throughput, backplane capability, packet loss rate, delay and delay jitter, burst processing capability, routing table capacity, quality of service, network management capability, reliability and availability. The packet loss rate is usually one of the performance indicators to measure the router when it is overloaded. The backplane capacity determines the throughput of the router. Voice and video services have high requirements on delay and jitter. Burst handling capability is measured by the maximum sending rate at which data packets can be sent at the smallest frame interval without causing loss. Therefore option A is correct.
Question 2: Which of the following descriptions about router technical indicators is wrong ( ).
A. Throughput refers to the packet forwarding capability of the router
B. The backplane capability determines the throughput of the router
C. Voice and video services have high requirements on delay and jitter
D. Burst processing capability is measured by the minimum frame interval value
Correct answer: D
reference analysis:
[Analysis] The burst processing capability of a router is measured by the maximum sending rate at which data packets can be sent at the minimum frame interval without causing loss, not just the minimum frame interval value. Therefore, option D is wrong.
Correct answer: D
Reference analysis:
[Analysis] The key technical indicators of the router mainly include the following points.
① Throughput: Throughput refers to the packet forwarding capability of the router. The packet forwarding capability of a router is related to the number of router ports, port speed, packet length, and packet type.
②Backplane capability: The backplane is the physical channel between the router input and output. The backplane capacity determines the throughput of the router.
③Packet loss rate: The packet loss rate refers to the probability of packet loss due to the limitation of packet forwarding capability under stable and continuous load conditions. The packet loss rate is usually one of the performance indicators to measure the router when it is overloaded.
④ Delay and delay jitter: Delay refers to the time from the first bit of the data packet entering the router to the last bit of the frame leaving the router. This time interval marks the processing time for the router to forward the packet. Delay jitter is the amount of variation in delay. Because data packets do not have high requirements on delay jitter, delay jitter is usually not used as the main indicator for measuring high-speed routers, but voice and video services have higher requirements on delay jitter.
⑤Burst processing capability: The burst processing capability is measured by the maximum sending rate at which data packets are sent at the minimum frame interval without causing loss. So choose option D
Question 4: Among the following descriptions about router technical indicators, the wrong one is ().
A. The packet forwarding capability of a router is related to the number of ports, port speed, packet length and packet type
B. High-performance routers generally use a shared backplane structure
C. The packet loss rate is one of the indicators to measure the router's overload capacity
D. The quality of service of the router is mainly manifested in the queue management mechanism and the type of QoS protocol supported
Correct answer: B
Refer to analysis:
[Analysis] The packet forwarding capability of a router is related to the number of ports, port rate, packet length and packet type. The packet loss rate is one of the indicators to measure the router's overload capacity. Traditional core routers use a shared backplane structure. The quality of service of the router is mainly reflected in the queue management mechanism and the type of QoS protocol supported高性能路由器一般采用采用可交换式的结构. Option B is wrong, so choose option B.
Question 13: Which of the following descriptions about router technical features is correct ( ).
A. High-performance routers generally use a shared backplane structure
B. The quality of service of the router is mainly reflected in the packet forwarding efficiency
C. The packet loss rate is one of the indicators to measure the router's overload capacity
D. The packet forwarding capability of the router is related to the number of ports, port type, packet length and packet type
Correct answer: C
Reference analysis:
[Analysis] High-performance routers generally adopt a switching structure. The quality of service of the router is mainly manifested in the queue management mechanism, port hardware queue management] and support for QoS protocols. The packet forwarding capability of a router is related to the number of ports, port speed, packet length, and packet type. So choose option C.
Question 8: Which of the following descriptions about router technical indicators is wrong ( ).
A. The packet forwarding capability of a router is related to the number of ports, port speed, packet length and packet type
B. The packet loss rate is one of the indicators to measure the router's overload capacity
C. The quality of service of the router is mainly reflected in the packet forwarding efficiency
D. High-performance routers generally adopt a switched structure
Correct answer: C
Reference analysis:
[Analysis] The quality of service of the router is mainly manifested in the queue management mechanism, port hardware queue management and support for QoS protocols.
Question 9: Which of the following descriptions about router technical indicators is wrong ().
A. Throughput refers to the packet forwarding capability of the router
B. The number of ports determines the throughput of the router
C. Voice and video services have high requirements on delay and jitter
D. The router determines the packet forwarding path through the routing table
Correct answer: B
Reference analysis:
[Analysis] The throughput refers to the packet forwarding capability of the router. The throughput of a router involves two aspects: port throughput and machine throughput. The port throughput refers to the packet forwarding capability of a specific port of the router, while the whole machine throughput refers to the packet forwarding capability of the whole router. The backplane is the physical pathway between routing inputs and outputs.背板能力决定了路由器的吞吐量. So choose option B
Question 12: Among the following descriptions about the technical characteristics of routers, the correct one is ().
A. Throughput refers to the routing table capacity of the router
B. The backplane capability determines the throughput of the router
C. Voice and video services have lower requirements for delay and jitter
D. Burst processing capability is measured by the minimum frame interval value
Correct answer: B
Reference analysis:
[Analysis] Among the key technical indicators of the router, the throughput refers to the packet forwarding capability of the router, and the backplane capability determines the throughput of the router. Voice and video services have higher requirements for delay jitter, while ordinary data packets do not have high requirements for delay jitter. Burst handling capability is measured by the maximum sending rate at which data packets can be sent at the smallest frame interval without causing loss. So choose option B.
Question 10: Which of the following descriptions about router technical indicators is correct ( ).
A. High-performance routers generally use a shared backplane structure
B. The capacity of the routing table is one of the indicators to measure the overload capacity of the router
C. The packet forwarding capability of the router is related to the number of ports, port type, packet length and packet type
D. The router's queue management mechanism refers to the router's queue scheduling algorithm and congestion management mechanism
Correct answer: D
Reference analysis:
[Analysis] Traditional routers generally adopt a shared backplane structure, and high-performance routers generally adopt a switched structure. The packet loss rate is one of the indicators to measure the router's overload capacity, and the routing table capacity indicator indicates the maximum number of routing table entries that the router can store. The packet forwarding capability of a router is related to the number of ports, port speed, packet length, and packet type. So choose option D.
Question 11: Among the following descriptions about router technical indicators, the correct one is ().
A. The number of ports determines the throughput of the router
B. Throughput refers to the routing table capacity of the router
C. Voice and video services have high requirements on delay and jitter
D. Burst processing capability is measured by the minimum frame interval value
Correct answer: C
Reference analysis:
[Analysis] The throughput refers to the packet forwarding capability of the router. The backplane is the physical pathway between routing inputs and outputs. The backplane capacity determines the throughput of the router. Since data packets do not have high requirements on delay jitter, delay jitter is usually not considered as the main index for measuring high-speed routers, but voice and video services have higher requirements on delay jitter. Burst handling capability is measured by the maximum sending rate at which data packets can be sent at the smallest frame interval without causing loss. So choose option C
Question 7: The working mode that must be entered to execute the ip route command of the router is ( )
A. User mode
B. Privileged mode
C. Routing protocol configuration mode
D. Global configuration mode
Correct answer: D
Reference analysis:
[Analysis]
The working modes of the router are:
1. User mode, when the Console logs in to the router, enter the correct password to enter the user mode, which is only a read-only mode. In user mode, only simple commands such as ping, telnet and show version can be executed.
2. In privileged mode, enter enable and the super user password to enter privileged mode. In privileged mode, you can view and save configuration files, clear flash memory, process and complete routing cold start and other operations. Routing protocol configuration mode, used for detailed configuration of routing protocols.
3. In the global configuration mode, you can configure the router’s host name, super user password, TFTP server, static route (ip route), access control list, multicast, IP accounting, etc., so the answer is D.
3. Routing protocol between autonomous system and Internet
1. Autonomous system
The routing of the Internet can only be carried out in a layered manner. First, the Internet is divided into multiple smaller autonomous systems by different management agencies.
An autonomous system (Autonomous System, AS) is a group of routers that share similar routing policies and operate under the same management organization. An AS can be a collection of routers running a single routing protocol, or it can be a collection of routers running different routing protocols, but all belong to an organization.
Its main feature is that it has the right to decide what routing strategy is used in the system. Routing within an autonomous system is called intra-domain routing, and routing between autonomous systems is called inter-domain routing.
The route used in the network inside the autonomous system is called the first layer of area routing, that is, the intra-area route, which is generated by the router inside the autonomous system and is responsible for the data transmission of the internal network. The second-layer route is the backbone route of the Internet. Each autonomous system connected to the Internet has a router responsible for connecting to the backbone network. They are responsible for maintaining the routing information between autonomous systems and responsible for data transmission between autonomous systems. When internal hosts in an autonomous system access external hosts, they need to pass through the backbone routers in the autonomous system.
2. Classification of Routing Protocols
Routing Protocol is simply a protocol used to calculate and maintain routing information. Routing protocols generally work at the transport layer or application layer of the OSI reference model.
Routed Protocol, also known as routed protocol, refers to a protocol that can be routed between different logical network segments by routers. Routing protocols usually work at the network layer of the OSI reference model.
According to the different scope of work, routing protocols can be divided into two categories: IGP (Interior Gateway Protocols, Interior Gateway Protocol) and EGP (Exterior Gateway Protocols, Exterior Gateway Protocol), as shown in the figure.
IGP and EGP
IGP is a routing protocol for exchanging routing information within the same autonomous system. The main purpose of IGP is to discover and calculate routing information inside the autonomous system:
- Routing Information Protocol (RIP) , which is a distributed, distance-vector-based routing protocol. ·
- Open Shortest Path First (OSPF) , which is a link-state routing protocol, is currently the most commonly used interior gateway protocol.
- Intermediate System to Intermediate System (IS-IS) , which provides dynamic routing for packets sent between two routers within a domain. ·
- Interior Gateway Routing Protocol (IGRP) , which is a distance vector routing protocol, is a proprietary protocol of Cisco.
EGP is used to connect different autonomous systems and exchange routing information between different autonomous systems. The main purpose of EGP is to transfer routing information from one autonomous system to another.
4. Interior Gateway Protocol
1. The basic concept of RIP
RIP (Routing Information Protocol) is a distributed, distance-vector-based interior gateway protocol that uses "hop count" to measure the routing distance to the destination address.
2. The principle of RIP
RIP stipulates that the router broadcasts a (V, D) message every 30 seconds , and the message information comes from the local routing table. In the (VD) message of the RIP protocol, the distance is calculated by the number of hops: the router directly connected to the target network is defined as one hop, and the distance between a router is two hops, and so on. The distance of a path is the number of routers on the path (from the source host to the destination host). RIP stipulates that the length of a finite path must not exceed 15, and exceeding 15 means that the path is infinitely long, that is, the path does not exist.
3. The operation process of RIP
(1) The establishment of the routing table.
The initialized router only contains all the network routes directly connected to the router, and the distance of each route in the (V, D) table is 0.
(2) Update of routing table information.
As shown in the figure is the update process of the routing table information, assuming that Router1 and Router2 are two adjacent routers in an autonomous system. Router2 sends (V, D) message to Router1, and then Router1 updates the routing table information according to the following rules.
If there is no record of this item in the routing table of Router1, Router1 adds this item in the routing table, and because it needs to be forwarded by Router2, the value of distance D is increased by 1; if a record in the routing table of Router1 is greater than the value of distance D minus 1 of a record sent by Router2, Router1 modifies this item in the routing table, and the value of distance D is increased by 1 according to the value provided by Router2.
Router 1 initial routing table
| destination network | distance | routing |
|---|---|---|
| 10.0.0.0 | 0 | direct |
| 20.0.0.0 | 8 | Router2 |
| 30.0.0.0 | 3 | Router2 |
| 120.0.0.0 | 11 | Router4 |
| 125.0.0.0 | 4 | Router5 |
| 212.0.0.0 | 10 | Router6 |
| 220.0.0.0 | 9 | Router6 |
Router 2 sends pair 1 (V, D) message
| destination network | distance |
|---|---|
| 10.0.0.0 | 3 |
| 20.0.0.0 | 4 |
| 30.0.0.0 | 2 |
| 40.0.0.0 | 7 |
| 120.0.0.0 | 5 |
Modified router 1 initial routing table
| destination network | distance | routing |
|---|---|---|
| 10.0.0.0 | 0 | direct |
| 20.0.0.0 | 5 | Router2 |
| 30.0.0.0 | 3 | Router2 |
| 40.0.0.0 | 8 | Router2 |
| 120.0.0.0 | 6 | Router2 |
| 125.0.0.0 | 4 | Router5 |
| 212.0.0.0 | 10 | Router6 |
| 220.0.0.0 | 9 | Router6 |
Update process of routing table information
4. RPR (Resilient Packet Ring) related real questions over the years
Relevant knowledge points of RPR (Resilient Packet Ring):
1. RPR is used like FDDI双环结构。
2. Every node in the RPR ring executes SRP 公平算法（不是 DPT、MPLS).
3. In the traditional FDDI ring, after 源结点successfully sending a data frame to the destination node, the data frame is 源结点recovered from the ring. But for RPR rings, this data frame is 目的结点recycled from the ring.
4. RPR adopts 自愈环设计ideas, can 50ms （不是30ms）isolate faulty nodes and fiber segments within a short time, and provide SDH-level fast protection and recovery.
5. RPR can 对不同的业务数据分配不同的优先级be used to efficiently transmit IP packets directly on optical fibers.
6. The maximum length of bare fiber between two RPR nodes can be reached 100公里.
7. Both of RPR 外环(顺时针)and 内环(逆时针)can use 统计复用the method of transmitting packets and control packets (not frequency division multiplexing).外顺内逆
Question 1: In the following description about RPR technology, 错误的是().
A、 RPR与FDDI—样使用双环结构
B、在RPR环中，源节点向目的节点成功发出的数据帧要由目的节点从环中收回
C、RPR环中每一个节点都执行 MPLS公平算法
D、RPR环能够在50ms内实现自愈
正确答案：C
参考解析:
【解析】弹性分组环(RPR)中每一个节点都执行SRP公平算法,与FDDI一样使用双环结构。传统的FDDI环中,当源结点向目的结点成功发送一个数据帧之后，这个数据帧要由源结点从环中回收，而RPR环限制数据帧只在源结点与目的结点之间的光纤段上传输，当源结点成功发送一个数据帧之后，这个数据帧由目的结点从环中回收。RPR采用自愈环设计思路，能在50ms时间内，隔离出现故障的结点和光纤段，提供SDH级的快速保护和恢复。
A、可以对不同的业务数据分配不同的优先级
B、能够在 100ms 内隔离出现故障的节点和光纤段
C、内环和外环都可以用于传输数据分组和控制分组
D、是一种用于直接在光纤上高效传输IP分组的传输技术
正确答案：B
参考解析：
【解析】RPR采用自愈环的设计思想，能够在50ms的时间内隔离出现故障的结点和光纤段，提供SDH级的快速保护和恢复，同时不需要像SDH那样必须有专用的带宽，因此又进一步提高了环带宽的利用率。故选项B错误。
A、RPR与FDDI—样使用双环结构
B、在RPR环中,源节点向目的节点成功发出的数据帧要由源节点从环中收回
C、 RPR环中每一个节点都执行SRP公平算法
D、 RPR能够在50ms内隔离出现故障的节点和光纤段
正确答案：B
参考解析:
【解析】 RPR数据帧要由目的结点从环中收回,并非源结点,从而不再占用下游段的环带宽,提高了环带宽的利用率，因此B选项错误。
A、RPR能够在50ms内隔离出现故障的节点和光纤段
B、RPR环中每一个节点都执行SRP公平算法
C、两个RPR节点之间的裸光纤最大长度为100公里
D、RPR用频分复用的方法传输IP分组
正确答案：D
参考解析:
【解析】 RPR将沿顺时针传输的光纤环叫做外环,将沿逆时针传输的光纤环叫做内环。内环和外环都可以用统计复用的方法传输IP分组。D选项错误。
A、RPR环中每一个节点都执行SRP公平算法
B、 在RPR环中，源节点向目的节点成功发出的数据帧要由源节点从环中收回
C、RPR将沿逆时针传输的光纤环叫做内环
D、RPR环能够在50ms内实现自愈
正确答案：B
参考解析：
【解析】弹性分组环（RPR）是一种直接在光纤上高效传输IP分组的传输技术，它采用双环结构，这一点与FDDI】结构相同,并将沿顺时针传输的光纤叫做外环,将沿逆时针传输的光纤叫做内环。RPR环限制数据帧只在源结点与目的结点之间的光纤段上传输,当源结点成功地发送一个数据帧之后,这个数据帧要由目的结点从环中收回。RPR环中每个结点都执行SRP公平算法，使得结点之间能够获得平等的带宽，防止个别结点因流量大而造成环拥塞。此外，RPR采用自愈环的设计思想，能够在50 ms的时间内隔离出现故障的结点和光纤段，提供SDH级快速保护和恢复。因此，B项表述错误。
第2题：下列关于RPR技术的描述中，正确的是（ ）。
A、 RPR环能够在30ms内实现自愈
B、在RPR环中,源节点向目的节点成功发出的数据帧要由源节点从环中收回
C、两个RPR节点之间的裸光纤最大长度可以达到100公里
D、RPR的内环用于传输数据分组，外环用于传输控制分组
正确答案：C
参考解析:
【解析】弹性分组环(RPR)采用自愈环设计思路,能在50ms时间内,隔离出现故障的结点和光纤段,提供SDH级的快速保护和恢复。RPR限制数据帧只在源结点与目的结点之间的光纤段上传输，当源结点成功发送一个数据帧之后，这个数据帧由目的结点从环中回收。两个RPR结点之间的裸光纤最大长度可以达到100km, RPR将顺时针传输方向的光纤环称为外环,将沿逆时针传输方向的光纤环称为内环,内环和外环都可以用统计复用的方法传输数据分组和控制分组。故C选项正确。
五、开放最短路径优先协议
1.OSPF协议概述
开放最短路径优先(Open Source Path First, DSPF)协议,也是内部网关协议的一种,主要用于在自治系统内部路由器之间传输路由信息,是基于Dijkstra提出的最短路径算法规则计算路由。
-
OSPF协议的主要特点有如下几点：OSPF协议最主要的特征就是使用分布式链路状态协议（Link StateProtocol)。
-
OSPF协议要求当链路发生变化时用洪泛法向本自治系统中所有路由器发送信息。
主要是路由器通过所有输出端口向所有相邻的路由器发送信息,而每个相邻路由器又再将此信息发往其所有的相邻路由器,最终所有在这个区域中的路由器都拥有了一个一致的网络结构信息。而RIP的每个路由器只知道到每个网段的距离和下一跳,不知道全网的拓扑结构,并且RIP仅向自己相邻的几个路由器交换路由信息。 -
OSPF发送的信息是与本路由器相邻的所有路由器的链路状态,所谓链路状态指的是本路由器相邻路由器都有哪些,以及与相邻路由器之间链路的“度量”值, OSPF的链路状态"度量”主要是指距离、费用、带宽、时延等。
2.OSPF的层次结构与区域划分
OSPF协议将自治系统划分成不同的区域。划分区域是从逻辑上将路由器划分为不同的组,每个组用一个区域号(Area ID)来标识。 划分区域的边界是路由器，而不是链路。一个网段（链路）只能属于一个区域，或者说运行OSPF路由器的每个接口必须属于某个特定的区域。区域号是0，通常被称为骨干区域或骨干路由器。 骨干区域负责区域之间的路由，非骨干区域之间的路由信息必须通过骨干区域来转发。
3.OSPF协议的运行过程
(1)路由器的初始化过程。
OSPF中的每个路由器都采用一个包含数据库描述的数据包,与相邻的路由器交换自己数据库中已有的链路状态摘要信息,而不是把自己的链路状态信息对全网广播。
(2)网络运行过程。
在网络运行的过程中,只要一个路由器的链路状态发生变化,该路由器就要使用链路状态更新分组,用洪泛法向全网更新链路状态。OSPF使用的是可靠的洪泛法。 为了确保链路状态数据库与全网的数据库状态保持一致，OSPF还规定每隔一段时间，例如每30分钟要刷新一次数据库中的链路状态。
4.OSPF相关历年真题例题
OSPF的相关知识点：
1、OSPF 是 内部网关协议 的一种,采用最短路径算法，使用 分布式链路状态协议。
2、对于规模很大的网络， OSPF通过 划分区域 来提高路由更新收敛速度。每个区域有一个 32 位的区域标识符，区域内路由器不超过 200 个。
3、一个 OSPF 区域内每个路由器的链路状态数据库包含着 本区域(不是全网)的拓扑结构信息,不知道其他区域的网络拓扑。
4、链路状态“度量”主要指费用、距离、延时、带宽 等,没有路径。
5、当链路状态发生变化时用 洪泛法 向 所有 (不是相邻)路由器发送此信息。
6、链路状态数据库中保存的是全网的拓扑结构图,而非一个完整的路由表,也不是只保存下一跳路由器的数据。
第 1 题：下列关于OSPF协议的描述中，错误的是（）。
A、对于规模很大的网络，OSPF通过划分区域来提高路由更新收敛速度
B、每一个OSPF区域拥有一个32位的区域标识符
C、在一个OSPF区域内部的路由器不知道其他区域的网络拓扑
D、在一个区域内的路由器数一般不超过24个
正确答案：D
参考解析:
【解析】为了适应大规模的网络,并使更新过程收敛的很快, OSPF协议将一个自治系统划分为若干个更小的范围,每个范围叫做区域。每个区域有一个32位的区域标识符(点分十进制表示) ,在一个区域内的路由器数目不超过200个。划分区域的好处是将利用洪泛法交换链路状态信息的范围局限在每一个区域内,而不是整个自治系统,因此区域内部路由器只知道本区域的完整网络拓扑，而不知道其他区域的网络拓扑情况。
A、对于规模很大的网络, OSPF通过划分区域不能提高路由更新收敛速度
B、每一个区域OSPF拥有一个32位的区域标识符
C、在一个OSPF区域内部的路由器不知道其他区域的网络拓扑
D、在一个区域内的路由器数一般不超过200个
正确答案：A
参考解析:
【解析】对于规模很大的网络, OSPF通过划分区域来提高路由更新收敛速度,每一个区域OSPF拥有一个32位的】区域标示符，在一个OSPF区域内部的路由器不知道其他区域的网络拓扑，在一个区域内的路由器数一般不超过200个。故选择A选项。
A、每一个OSPF区域拥有一个32位的区域标识符
B、 OSPF区域内每个路由器的链路状态数据库包含着全网的拓扑结构信息
C、 OSPF协议要求当链路状态发生变化时用洪泛法发送此信息
D、距离、延时、带宽都可以作为OSPF协议链路状态度量
正确答案：B
参考解析:
【解析】 OSPF区域内每个路由器的链路状态包含着本区域而不是全网的拓扑结构信息。每一个OSPF区域拥有一个32位的区域标识符,在一个区域内的路由器数不超过200, OSPF协议要求当链路发生变化时用洪泛法向所有路由器发送此信息。链路状态“度量”主要是指费用、距离、延时、带宽等，综上所述B错。
A、OSPF使用分布式链路状态协议
B、链路状态“度量”主要是指费用、距离、延时、带宽等
C、当链路状态发生变化时用洪泛法向所有路由器发送信息
D、 链路状态数据库只保存下一跳路由器的数据
正确答案：D
参考解析：
【解析】OSPF使用分布式链路状态协议，链路状态“度量"主要是指费用、距离、延时、带宽等。当链路状态发生变化时用洪泛法向所有路由器发送此信息。使用OSPF协议的路由器都建立一个链路状态数据库,即为全网的拓扑结构图,而D项中只保存下—跳路由器数据的是RIP协议。因此，D选项错误。
A、 OSPF使用分布式链路状态协议
B、链路状态协议“度量”主要是指费用、距离、延时、带宽等
C、当链路状态发生变化时用洪泛法向所有路由器发送信息
D、链路状态数据库中保存一个完整的路由表
正确答案：D
参考解析:
【解析】最短路径优先协议(OSPF)最主要的特征是使用分布式链路状态协议。它要求路由器发送分信息是本路】由器与路由器相邻，以及链路状态的度量。OSPF协议要求当链路状态发生变化时使用洪泛法向所有路由器发送消息。由于执行OSPF协议的路由器之间频繁地交换链路状态信息，因此所有的路由器最终都能建立一个链路状态数据库。该数据库实际上就是全网的拓扑结构图，并且在全网范围内是保持—致的。
本题D选项说法有误，链路状态数据库中保存的是全网的拓扑结构图，而非全网完整的路由表。
A、对于规模很大的网络，OSPF通过划分区域来提高路由更新收敛速度
B、每一个区域OSPF拥有一个30位的区域标示符
C、在一个OSPF区域内部的路由器不知道其他区域的网络拓扑
D、在一个区域内的路由器数一般不超过200个
正确答案：B
参考解析:
【解析】对于规模很大的网络, OSPF通过划分区域来提高路由更新收敛速度,每一个区域OSPF拥有一个32位的区域标示符,在一个OSPF区域内部的路由器不知道其他区域的网络拓扑,在一个区域内的路由器数一般不超过200个。由解析知B选项错误，故选择B选项。
第 2 题：每一个OSPF区域拥有一个区域标识符，区域标识符的位数是 （ ）。
A、16
B、32
C、48
D、64
正确答案：B
参考解析:
【解析】每一个OSPF区域拥有一个32位的区域标识符,每一个区域内的路由器数不超过200个。
第 3 题：下列关于路由选择协议相关技术的描述中，错误的是（）。
A、最短路径优先协议使用分布式链路状态协议
B、路由信息协议是一种基于距离向量的路由选择协议
C、链路状态度量主要包括带宽、距离、收敛时间等
D、边界网关协议可以在两个自治域系统间传递路由选择信息
正确答案：C
参考解析:
【解析】最短路径优先协议(OSPF)最主要的特征是使用分布式链路状态协议。它要求路由器发送分信息是本路由器与路由器相邻,以及链路状态的度量。链路状态的度量主要有带宽、距离、费用、延时等。收敛时间是RIP距离 向量的度量参数之一。路由信息协议(RIP)是内部网关协议中使用最广泛的一种协议,它是一种分布式、基于距离向量的路由选择协议。外部网关协议（EGP）是不同自治系统的路由器之间交换路由信息的协议，其典型协议是边界网关协议（BGP）。
A、OSPF采用的是一种链路状态算法
B、OSPF可以划分区域，路由更新信息可在不同区域间传播
C、区域用数字标识(ID) ,区域ID是一个32位无符号数值
D、ID可以用十进制整数和点分十进制数表示形式
正确答案：B
参考解析:
【解析】 OSPF可以划分区域,路由更新信息在同一个区域内传播。区域边界路由器负责接收从其他区域来的信息。
第 4 题：下列关于路由协议的描述中，正确的是（）。
A、 OSPF协议中，链路状态数据库中保存一个完整的路由表
B、OSPF协议中，链路状态协议“度量”主要是指距离和延迟
C、 RIP协议中,路由刷新报文主要内容是由若干(D、V)组成的表
D、RIP协议中，距离D标识该路由器可以到达的目的网络或目的主机的跳数
正确答案：D
参考解析: D
【解析】 OSPF使用分布式链路状态协议,链路状态数据库存储着全网的拓扑结构图,并且在全网内保持一致,链路状态协议度量"主要是指费用、距离、延时、带宽等。RIP协议中,路由刷新报文主要内容是由若干(V、 D)组成的表,矢量v标识该路由器可以到达的目的网络或目的主机，距离D标识该路由器可以到达的目的网络或目的主机的跳数。
A、OSPF协议中，链路状态数据库中保存一个完整的路由表
B、OSPF协议中，链路状态协议“度量”是指距离、带宽
C、RIP协议中,路由刷新报文主要内容是由若干(V、D)组成的表
D、RIP协议中，矢量V标识该路由器可以到达的目的网络或目的主机的跳数
正确答案：C
参考解析：
【解析】OSPF使用分布式链路状态协议，链路状态数据库存储着全网的拓扑结构图，并且在全网范围内保持一致,链路状态协议"度量"主要是指费用、距离、延时、带宽等。RIP协议中,路由刷新报文主要内容是由若干(V、D)组成的表,矢量V标识该路由器可以到达的目的网络或目的主机，矢量D标识该路由器可以到达的目的网络或目的主机的跳数。故选择C选项。
A、RIP协议中，路由刷新报文主要内容是由若干(D、V)组成的表
B、RIP协议中，路由器在接收到更新报文后按照最短路径原则更新路由表
C、OSPF协议中，链路状态数据库中保存一个完整的路由表
D、OSPF协议中，当链路状态发生变化时用洪泛法向相邻路由器发送信息
正确答案：B
参考解析: B
【解析】 RIP协议中,路由刷新报文主要内容是由若干(V、 D)组成的表,矢量V标识该路由器可以叨叨的目的网络或目的主机，矢量D标识该路由器可以到达的目的网络或目的主机的跳数。OSPF使用分布式链路状态协议，链路状态数据库存储着全网的拓扑结构图,并且在全网内保持一致,链路状态发生变化时用洪泛法向是所有路由器发送信息。
A、RIP协议适用于路由变化剧烈的互联网络环境
B、RIP协议中,路由器在接收到更新报文后按照最短路径原则更新路由表
C、 OSPF协议中,链路状态协议"度量"是指距离、带宽
D、 OSPF协议中，每一个区域OSPF拥有一个128位的区域标识符
正确答案：B
参考解析：
【解析】RIP协议是一种分布式、基于距离向量的路由选择协议，其优点是实现简单，但是它不适用于大型或路由变化剧烈的互联网环境。RIP协议中，路由刷新报文主要内容是由若干(V,D)组成的表,矢量V标识该路由器可以到达的目的网络或目的主机,矢量D标识该路由器可以到达的目的网络或目的主机的跳数。其它路由器在接收到某个路由器的(V, D)报文后，按照最短路径原则对各自的路由表进行刷新。OSPF使用分布式链路状态协议，链路状态数据库存储着全网的拓扑结构图,并且在全网内保持一致,链路状态协议"度量"主要是指费用、距离、延时、带宽等, OSPF协议中,每一个区域OSPF拥有一个32位的区域标识符。故选择B选项。
A、外部网关协议是不同自治系统的路由器之间交换路由信息的协议
B、 RIP协议适用于路由变化剧烈的互连网络环境
C、与RIP相比, OSPF协议的"度量"更侧重于距离和带宽
D、 OSPF协议中，每一个区域OSPF拥有一个128位的区域标示符
正确答案：A
参考解析:
【解析】外部网关协议是不同自治系统的路由器之间交换路由信息的协议。RIP协议是一种分布式、基于距离向量的路由选择协议，其优点是实现简单，但是它不适用于大型或路由变化剧烈的互联网环境。OSPF使用分布式链路状态协议，链路状态协议“度量”主要是指费用、距离、延时、带宽等，OSPF协议中，每一个区域OSPF拥有一个32位的区域标识符。故选择A选项。
第5 题:下列关于配置OSPF的描述中,错误的是（ ）。
A、 OSPF区域的数字标识称为区域ID
B、骨干区域的ID为0
C、区域ID为0的另一种表示形式是0.0.0.0
D、区域ID的取值范围是0-200
正确答案：D
参考解析：
【解析】 OSPF区域的数字标识称为区域ID，区域ID是一个32位无符号数，数值范围是0~4294967295，区域ID为0的另一种表示形式是0.0.0.0.如果区域ID为0或0.0.0.0时,则表示骨干区域,因此D选项错误,故选择D选项。
第 6 题：下列关于OSPF协议的描述中，正确的是（）。
A、在一个区域内的路由器数可达300个
B、每一个区域OSPF拥有一个30位的区域标识符
C、一个OSPF区域内的路由器不知道其他区域的网络拓扑
D、当链路状态发生变化时只需向相邻的路由器发送信息
正确答案:C
参考解析:
【解析】对于规模很大的网络, OSPF通过划分区域来提高路由更新收敛速度,每一个区域OSPF拥有一个32位的区域标识符,在一个OSPF区域内部的路由器不知道其他区域的网络拓扑,在一个区域内的路由器数一般不超过200个。当链路状态发生变化时用洪泛法向所有的路由器发送信息。故选C选项。
A、 OSPF使用距离向量算法
B、链路状态数据库中包含所有区域的拓扑结构图
C、链路状态“度量”主要是指费用、距离、延时、带宽等
D、当链路状态发生变化时只需向相邻的路由器发送信息
正确答案：C
参考解析:
【解析】 OSPF采用最短路径算法,其链路状态数据库(link state database)包含全网的拓扑结构图。当链路状态发送变化时使用洪泛法向所有路由器发送信息。故选C选项。
A、链路状态协议“度量”用跳数表示
B、链路状态数据库中保存一个完整的路由表
C、每一个OSPF区域拥有一个30位的区域标识符
D、当链路状态发生变化时用洪泛法向所有路由器发送信息
正确答案：D
参考解析:
【解析】 OSPF中链路状态协议的度量主要指费用、距离、延时、带宽等。链路状态数据库中保存一个拓扑结构图，并且在全网内保持一致。每一个区域有一个32位的区域标识符，在一个区域内的路由器数不超过200个。当链路状态发生变化时用洪泛法向所有路由器发送信息。
A、OSPF使用分布式链路状态协议
B、链路状态数据库只保存下-跳路由器的数据
C、链路状态“度量”主要是指距离和延时
D、为确保链路状态数据库一致, OSPF每隔30秒刷新一次数据库中的链路状态
正确答案：A
参考解析:
【解析】 OSPF协议最主要的特征是使用分布式链路状态协议(link state protocol) ,其度量主要指费用、距离、延时、带宽等。由于执行OSPF协议的路由器之间频繁地交换链路信息,因此所有的路由器最终都能建立一个链路状态数据库，这个数据库实际上就是全网的拓扑结构图，并且在全网范围内是保持一致。每一个区域有一个32位的区域标识符，在一个区域内的路由器数不超过200个。当链路状态发生变化时用洪泛法(flooding)向所有路由器发送信息。故选择A选项。
A、链路状态数据库中保存一个完整的路由表
B、链路状态协议"度量"主要是指距离、带宽等
C、每一个区域OSPF拥有一个32位的区域标识符
D、当链路状态发生变化时使用 UDP协议 向相邻 路由器发送信息
正确答案：C
参考解析：
【解析】OSPF中链路状态协议的度量主要指费用、距离、延时、带宽等。链路状态数据库中保存一个全网的拓扑结构图，并且在全网范围内是保持一致。每一个区域有一个32位的区域标识符，在一个区域内的路由器数不超过200个。当链路状态发生变化时用洪泛法（flooding)向所有路由器发送信息。故选择C选项。
六、外部网关协议
1.BGP协议的特性
(1) BGP是一种外部网关协议,它重点关心的不是发现和计算路由,而是在AS之间传递路由信息以及控制优化路由信息(提供这些信息是为了找到较好的路由，不一定是最佳路由）。
(2) BGP是一种“路径矢量”路由协议，其路由信息中携带了所经过的全部自治系统的路径列表。
(3)为了保证BGP的可靠传输，其使用TCP来承载协议报文。 通过TCP的可靠传输机制、重传、排序等机制来保证BGP消息报文传输的可靠性。
(4) BGP支持CIDR和路由聚合，可以将一些连续的子网聚合成较大的子网(突破了标准分类地址限制),从而可以在一定程度上控制路由表的快速增长,并降低了路由查找的复杂度。
(5) BGP最重要的特性是丰富的路由属性以及强大的路由过滤和路由策略。
2.BGP的路由器角色及基本特征
(1)发言者。
发送BGP协议报文的路由器称为BGP发言者(BGP Speaker) ,它接受或产生新的路由信息,并发布给其他发言者。
(2)对等体。
相互交换BGP协议报文的BGP发言者之间互称为BGP对等体（BGPPeer)。
(3)IBGP对等体。
处于同一个AS (域)的BGP对等体称为IBGP对等体，从IBGP获得的路由不向它的IBGP对等体发布。图中RB、RD是IBGP对等体。
(4)EBGP对等体。
处于不同AS的BGP对等体称为EBGP对等体。通常情况下, EBGP对等体是物理上直连的BGP发言者,路由器从EBGP对等体获得的路由会向它所有的BGP对等体通告(包括EBGP和IBGP) 。
3.BGP协议报文
(1)打开(Open)分组: 打开分组用来与相邻的另一个BGP发言人建立关系。
(2)保活(Keepalive)分组: 保活分组用来确认打开报文，以及周期性地证实相邻边界路由器的存在。
(3)更新(Update)分组: 更新分组用来发送某一路由的信息,以及列出要撤销的多条路由。
(4)通知(Notification)分组： 通知分组的作用为发出错误通知。BGP发言者如果检测到对方发过来的消息有错误或者对方主动断开BGP连接,都会发出通知报文通知BGP邻居。
4.BGP的运行过程
当开始向相邻边界路由器进行协商时就要发送“打开分组”,如果相邻边界路由器接受,就响应一个“保活分组”。两个BGP发言人的相邻关系就建立了。为了维护这种关系,两个BGP发言人彼此要周期性地(一般是每隔30秒)交换“保活分组”。"更新分组”是BGP协议的核心, BGP发言人可以用“更新分组”撤销它以前曾经通知过的路由,也可以宣布增加新的路由。
5.BGP相关历年真题例题
BGP（边界网关协议）的相关知识点：
1、BGP 是边界网关协议，是外部而不是内部网关协议（是不同自治系统的路由器之间使用的协议）。
2、一个 BGP 发言人使用 TCP (而不是UDP)与其他自治系统的 BGP 发言人 交换路由信息。
3、BGP 交换路由信息的节点数是以自治系统数为单位的， BGP 交换路由信息的节点数不小于自治系统数。
4、BGP-4 采用路由向量协议，而 RIP 采用距离向量协议。
5、BGP 发言人通过update而不是 noticfication 分组通知相邻系统，使用 update 分组更新路由时，一个报文只能增加一条路由。
6、open 分组用来与相邻的另一个BGP发言人建立关系，两个 BGP 发言人需要 周期性地（不是不定期）交换 keepalive 分组来确认双方的相邻关系。
7、BGP 路由选择协议执行中使用的四个分组为 打开（open）、更新（update）、报活（keepalive）、和通知（notification） 分组。
第1题:下列关于外部网关协议BGP的描述中,错误的是()。
A, BGP是不同自治系统的路由器之间交换路由信息的协议
B、一个BGP发言人使用UDP与其他自治系统中的BGP发言人交换路由信息
C、BGP协议交换路由信息的节点数不小于自治系统数
D、BGP-4采用路由向量协议
正确答案：B
参考解析:
【解析】外部网关协议是不同的自治系统的路由器之间交换路由信息的协议。一个BGP发言人与其他自治系统中的BGP发言人要交换路由信息就要先建立TCP连接,然后在此连接上交换BGP报文以建立BGP会话。BGP协议交换路由信息的结点数是以自治系统数为单位的,所以不会小于自治系统数。BGP4采用路由向量路由协议。
第 2 题：下列关于BGP协议的描述中，错误的是（ ）。
A、 当路由信息发生变化时，BGP发言人使用 notification 分组通知相邻自治系统
B、一个BGP发言人与其他自治系统中BGP发言人交换路由信息使用TCP连接
C、 open分组用来与相邻的另一个BGP发言人建立关系
D.两个BGP发言人需要周期性地交换keepalive分组来确认双方的相邻关系
正确答案：A
参考解析:
【解析】路由信息发生变化时, BGP发言人通过update而不是notification分组通知相邻AS, Open报文用来与相邻的另一个BGP发言人建立关系。UPDATE 更新分组用来发送某一路由的信息，以及列出要撤销的多条路由。keepalive 保活分组用来确认打开的报文，以及周期性地证实相邻的边界路由器的存在。一个BGP发言人与其它自治系统中的BGP发言人要交接路由信息，就是先建立TCP连接，然后在此连接上交换BGP报文以建立BGP会话。因此选项A错误。
第 3 题：以下协议中不属于内部网关协议的是（ ）。
A、 RIP
B、OSPF
C、 BGP
D、IGRP
正确答案：C
参考解析:
【解析】内部网关协议主要有RIP和OSPF协议。BGP是边界网关协议,不是内部网关协议
第4题:不同AS之间使用的路由协议是( )。
A、BGP
B、ISIS
C、 OSPF
D、RIP
正确答案：A
参考解析:
【解析】外部网关协议是不同自治系统的路由器之间交换路由信息的协议。1989年发布了主要的外部网关协议–边界网关协议(BGP)。新版本BGP 4是在1995年发布的,并采用路由向量路由协议。中间系统到中间系统(IS IS,Intermediate system to intermediate system,读作1 sys")是一种内部网关协议,是电信运营商普遍采用的内部网关协议之一。故选项B不正确; OSPF(Open Shortest Path First开放式最短路径优先)是内部网关协议(Interior Gateway Protocol,简称IGP) ,用于在单一自治系统(autonomous system,AS)内决策路由。题目的要求是不同的自治系统,故选择C不正确;路由信息协议（RIP）是一种在网关与主机之间交换路由选择信息的标准。RIP 是一种内部网关协议。故选项D不正确。
A、RIP
B、OSPF
C、BGP-4
D、IS-IS
正确答案：C
参考解析:
【解析】 AS (Autonomous System,自治系统)是一个有权自主地决定在本系统中采用何种路由协议的小型单位。路由信息协议(RIP)是一种在网关与主机之间交换路由选择信息的标准, RIP是一种内部网关协议.OSPF(Open Shortest Path First开放式最短路径优先)是一个内部网关协议用于在单一自治系统内决策路由。 BGP是一种外部网关协议,用于不同自治系统的路由器之间交换路由信息的协议。IS-IS(Intermediate System to Intermediate System Routing Protocol,中间系统到中间系统的路由选择协议)是一种内部网关协议(IGP) ,是电信运营商普遍采用的内部网关协议之一。综上可知,不同自治系统之间的应当使用的是外部网关协议。C选项符合要求，故选C选项。
第 5 题：BGP路由选择协议的四种分组中不包括（）。
A、inform
B、keepalive
C、open
D、update
正确答案：A
参考解析:
【解析】 BGP路由选择协议执行中使用的四个分组为打开(open) 、更新(update) 、保活(keepalive)和通知（notification）分组。因此，A选项正确。
第 6 题：下列关于路由选择协议相关技术的描述中，错误的是（）。
A、最短路径优先协议使用分布式链路状态协议
B、路由信息协议是一种基于距离向量的路由选择协议
C、链路状态度量主要包括带宽、距离、收敛时间等
D、边界网关协议可以在两个自治域系统间传递路由选择信息
正确答案：C
参考解析:
【解析】最短路径优先协议(OSPF)最主要的特征是使用分布式链路状态协议。它要求路由器发送分信息是本路由器与路由器相邻，以及链路状态的度量。链路状态的度量主要有带宽、距离、费用、延时等。收敛时间是RIP距离向量的度量参数之一。路由信息协议(RIP)是内部网关协议中使用最广泛的一种协议，它是一种分布式、基于距离向量的路由选择协议。外部网关协议(EGP)是不同自治系统的路由器之间交换路由信息的协议,其典型协议是边界网关协议(BGP)
第7题:下列关于外部网关协议BGP的描述中,正确的是(）
A、一个BGP发言人使用UDP与其他自治系统中的BGP发言人交换路由信息
B、BGP协议交换路由信息的结点数是以自治系统数为单位的
C、BGP是同一自治系统的路由器之间交换路由信息的协议
D、BGP采用距离向量协议
正确答案：B
参考解析:
【解析】 BGP是不同自治系统的路由器之间交换路由信息的协议,一个BGP发言人使用TCP与其他自治系统中的BGP发言人交换路由信息, BGP协议交换路由信息的结点数是以自治系统数为单位的, BGP-4采用路由向量协议。
第 8 题：BGP协议的分组中，需要周期性交换的是（）。
A、open
B、update
C、keepalive
D、notification
Correct answer: C
Reference analysis:
[Analysis] BGP speakers need to periodically (usually every 30 seconds) exchange "keep alive packets (keeplive)". So choose option C.
Question 9: In the following description about the BGP protocol, 错误的是().
A. Two BGP speakers need to 不定期地exchange keepalive packets to confirm the adjacency between the two parties.
B. A BGP speaker exchanges routing information with BGP speakers in other autonomous systems using a TCP connection. C. The
open group is used to establish a relationship with another adjacent BGP speaker
. D. When updating routes with update packets, only one route can be added to a message
Correct answer: A
Reference analysis:
[Analysis] Two BGP speakers need to exchange keepalive packets periodically (usually every 30s) to confirm the neighbor relationship between the two parties. So choose option A.