[Computer Network] Computer Network Interview Questions series 2

[Computer Network] Computer Network interview series II 31. Telecommunication Network classification telecommunication network circuit exchange network group exchange network FDMTDM virtual circuit network datagram network 32. What is the network classification by region? Lan, man, and wan. 33. networks are classified by users as public networks and private networks. 34. The main network topologies are: Star, bus, ring, tree, full connection, and irregular mesh. 35. Computer Network Architecture? It is actually a set of protocols corresponding to the layer-by-layer addition. The Protocol consists of three components: Syntax: structure or format of data and control information; semantics: the control information to be sent, the action to be done, and the response to the request; timing (synchronization ): detailed description of event Implementation sequence. 36. What is the line pair of twisted pair wires? 1-2, 7-8, 3-6, 4-5 white-blue, white-orange, white-green, white-brown 37, data link layer protocols may provide service? Frame formation, Link Access, transparent transmission, reliable delivery, traffic control, error detection, error correction, half-duplex and full-duplex. The most important thing is frame demarcation (frame formation), transparent transmission, and error detection. 38. Frame demarcation? Frame boundary is used to determine the frame boundary. The methods include byte notation, character filling, and zero-bit filling. 39. transparent transmission? That is, any data should be transmitted. The starting and ending characters used to mark the frame in the frame demarcation should also be correctly transmitted. 40. error detection? Cyclic Redundancy checks CRC. The calculated result is the frame inspection sequence (FCS. The CRC error detection technology of the cyclic redundancy test sequence can only be accepted without errors, that is, all frames accepted at the data link layer at the receiving end, we can all think that these frames do not produce errors during transmission with a very close probability of 1, but reliable transmission is required (that is, what is sent), that is, frames transmitted to the receiving end are error-free, lost, and repeat-free, and are also received in the order of sending. In this case, confirmation and retransmission mechanisms must be added. 41. How can I implement reliable transmission protocols? 1) Stop waiting Protocol: stop sending after each frame is sent until the next frame is received and received. If no confirmation is received from the receiving end, then, the previous frame is retransmitted through the preset timer. There are three possibilities: retransmission may cause the receiver to receive the same frame. At this time, it is determined by the serial number. If the serial number of the received frame has been received before, the newly received frame is discarded. Because the receiving end may not be able to receive the message correctly at one time, frames must be backed up at the sending end until they are confirmed, because the Protocol can only send one frame at a time, therefore, the sending cache does not need to be too large. 2) continuous ARQ protocol: The sending window is greater than 1 and the receiving window is equal to 1. Therefore, the sending window has been sent to the frame with the serial number 5, however, when an error occurs when the receiving end receives a frame with the serial number 3, all frames after the 3rd must be retransmitted. Therefore, an error may cause retransmission of multiple frames, at the same time, in order to be able to re-transmit when an error occurs, the frames that have not been confirmed to be sent must be saved in the transmission end's full-latency zone, in this case, the buffer zone is more required than the Stop wait protocol. However, when n bits are used to represent numbers, the protocol can work correctly only when the size of the sending window is. If n bits are used, the size of the sending window is WT <= 2n-1. 3) Select the retransmission ARQ protocol: when both the sending and receiving windows are greater than 1, this may reduce the number of retransmission frames. If n-bit numbers are used, the size of the receiving window is WR 00002n/2. 42. How does the PPP protocol work? When a user dials to access the ISP, the ISP's modem confirms the dialing and establishes a physical link. The user sends a series of LCP groups to the ISP's router. This selects some parameters for the PPP, then configure the network layer. NCP assigns a temporary IP address to the new connected PC, so that the user PC becomes a host on the Internet. After the communication ends, NCP releases the network layer connection to reclaim the IP address. Then, LCP releases the data link layer connection, and finally releases the physical layer connection. 43. data link layer interconnection equipment 1) Bridge: two interconnected networks adopt different data link layer protocols, different transmission media and networks with different transmission rates, and the network connected by the bridge adopts the same protocol above the data link layer. 2) Storage and forwarding devices connected by switches on the data link layer. The switch simply decides the information Forwarding Based on the MAC address in each packet. The switch corresponds to the hardware device and the software corresponding to the bridge. 44. What are the key technologies of lan? Topology (Star, bus, ring, tree) Media Access Mode (CSMA/CD, Token-passing) signal transmission mode (baseband, broadband ). 45. What are the functions of the network interface card (NIC? 1) perform serial/parallel conversion. 2) cache data. 3) install the device driver on the computer's operating system. 4) Implement the Ethernet protocol. 46. CSMA/CD? Answer: it refers to the Multi-Point Access/collision detection of the carrier listener (1) Multi-Point Access refers to the Multi-Point Access of multiple computers connected to one bus (2) carrier monitoring means that before sending data, each station must first check whether other computers are sending data on the bus. If so, do not send data temporarily to avoid collision (3) actually, there is no carrier on the bus. Actually, the electronic technology is used to check whether there are data signals sent by other computers on the bus (4) collision Detection is used to detect the signal voltage on the channel while the computer sends data. When a collision occurs, a conflict occurs. Collision Detection is also called "Conflict Detection" (5) when a collision occurs, the signal transmitted on the bus is distorted and useful information cannot be restored. Therefore, in order not to waste network resources, once a collision is detected, stop sending data. Then wait for a random period of time before sending. (6) After the collision is detected, the system not only immediately stops sending data, but also sends some interference information to other sites to know that the collision occurred. (7) Because signal transmission on the bus also takes some time, when a station detects that the bus is idle, it may not be truly idle, because there will be data sent by other sites, but it has not been transferred to the range that the site can detect. In this case, sending data will eventually lead to a collision. (8) Working Principle 1) monitor whether the channel is idle before sending, and send it immediately if it is idle; 2) if the channel is busy, continue to listen and send it immediately if it is idle; 3) you still need to listen during the sending process. If a conflict is detected, the system immediately stops sending data and sends a series of Interference Signals (Jam). 4) the Jam signal is sent to strengthen the conflict, so that all sites can detect a conflict. Wait for a random period of time (called backoff) and try again. In conclusion, there are four sentences: Listen before sending, send when idle, and send and listen while sending, so as to avoid conflict. 47. Ethernet MAC frame format? Destination Address (6 bytes) Source Address (6 bytes) type (2 bytes) data (46 -- 1500 bytes) FCS (4 bytes) the MAC address has a 48-bit type to identify what is used by the upper layer protocol, and the frame should be handed over to the upper layer. What protocol should I determine the end of The data field? After sending an Ethernet frame, the sender no longer sends other code elements, so the voltage on the network adapter interface of the sender does not change. Based on the end position, the first four bytes are the end position of the data field. 48. Virtual lan vlan? 1) VLAN limits the number of workstations that receive broadcast messages, so that the network will not deteriorate performance due to the spread of too much broadcast information (that is, broadcast storm. 2) VLAN Division: Based on ports, MAC addresses, and IP addresses. 3) VLAN frame format Destination Address (6 bytes) Source Address (6 bytes) VLAN tag (indicating which VLAN the site belongs to) type (2 bytes) data (46-1500 bytes) FC (4 byte) 49. What is the MAC layer of the wireless lan? 1) hiding site problems and exposing site problems 2) CSMA/CA: it is an improved CSMA/CD, and the added function is collision avoidance. In reality, it is to reserve a channel before sending data. 50. NAT? 1) Network Address Translation is a conversion technology that converts private addresses into valid IP addresses. This technology can solve the problem of insufficient IP addresses. 2) NAT implementation methods: static translation, dynamic conversion, port multiplexing (internal IP + port number-External IP + port number, in this way, the source port of the outgoing packet is changed and the port is switched. All hosts in the internal network can share a valid external IP address to access the Internet, thus saving IP resources, hides all hosts in the network to effectively prevent attacks from the Internet ). Disadvantage: Because the IP address in the IP Address Header needs to be converted, encryption operations cannot be performed for 51 or private (retained) addresses? Class A: 10.0.0. 0 -- 10.20.255.255 Class B: 172.16.0.0 -- 172.31.255.255 Class C: 192.168.0.0 -- 192.168.255.25552. What is the difference between switching and routing? What are the characteristics of a VLAN? Switching refers to forwarding and filtering frames. It is the work of a switch. In the OSI reference model, Layer 2 routing refers to a non-directly connected link in a network line. It is the work of a router, on the Layer 3 of the OSI reference model. 1) the IP address is not required for switching, but for routing, because the IP address is the layer-3 protocol and the layer-2 requires the MAC address. 2) the Layer 2 technology is different from the Layer 3 technology. The Layer 2 can be used for VLAN, port binding, and NAT, ACL, and QoS. VLAN, short for a virtual LAN, is a layer-2 technology that features three features: control broadcast, security, flexibility, and scalability. 53. SNMP? The abbreviation of Simple Network Management Protocol. 54. What is TTL? What is the role? Which tools will use it (ping, traceroute, ifconfig, netstat )? TTL refers to the survival time. In short, it indicates the time of the data packet in the network. After a router passes through, TTL is reduced by one, so that the TTL is eventually reduced to 0, when the TTL is 0, the data packet is discarded. This is because a ring may be formed between the two routers. If no TTL limit is set, the data packet will continue to die on this ring, when TTL is 0, the data packet is discarded. Ping sends data packets with TTL, but it is not necessary. That is, traceroute can work normally without TTL. ifconfig is used to configure Nic information, no TTL is required. netstat is used to display the route table and does not require TTL. 55. What is the use of the route table? In Linux, how does one configure a default route? The route table is used to determine how to transmit a data packet from a sub-network to another subnet. In other words, it is used to determine which Nic the packet received from a NIC should be sent. Each line in the route table must contain at least three information: the destination network number, subnet mask, and network adapter to which the subnet should be used. When a router receives a packet from a network adapter, It scans each line in the route table and uses the subnet mask and the destination IP address in the packet for logic and operation (&) find the target network number. If the resulting network number is the same as the network number of this line, the route table will be left as a backup route. If there is already a standby route, the longest network number will be left in the two routes and the other one will be discarded (this is the case where the classless addressing CIDR is the longest matching network number, in other cases, the first matching row can be found for forwarding ). Scan the next row until the end. If no route is found after scanning, use the default route. After the route is determined, the data packet is directly sent to the corresponding Nic. In specific implementation, the route table may contain more information for the details of the routing algorithm. On Linux, you can use the "route add default gw <default router IP address>" command to configure a default route. 56. Which five pieces of information do each router need to know when looking for a route? Destination Address: Determine the target address of the message sent from the neighboring station: specify who is directly connected to the router interface for route discovery: Find the neighboring station to know which network routes are selected: based on the information learned from the neighboring station, the optimal route to the destination is provided to keep the route information: the router saves a route table and stores all the route information it knows. 57. EGP, IGP? 1) IGP: the Internal Gateway Protocol, that is, the routing selection protocol used inside an autonomous system, such as RIP and OSPF. (1) RIP is a routing selection protocol based on distance vectors. Every vro in the network must maintain the distance vector from itself to every other destination network. Distance is the number of hops. The number of network hops directly connected to the vro is 1, and each hop passing through a vro increases by 1. A rip can contain up to 15 routers. If the number of routers is greater than 15, the path cannot be reached. The RIP can only work in a small network. Storage Structure: To a certain network distance, the next hop route will process the RIP packet sent from each adjacent Route step 1> receive the packet (destination network N, distance, next hop route ), the next hop is changed to X, and the distance is increased by 2> if there is no destination network N, the route table is updated. If the next hop is X, it must be updated regardless of the distance, because this is the latest message. If the distance in the received destination network is less than the original distance, the update table is not received for 3> 3 minutes, mark the adjacent router as the three key points of the non-reachable RIP: exchanging information with the adjacent router only; Exchanging information is all the information that the current router knows, that is, the route table; exchange route information at a fixed interval, such as 30 seconds. The RIP protocol uses the User Datagram UDP of the transport layer for transmission. Therefore, the location of the RIP Protocol is at the application layer, but the IP datagram forwarding process is completed at the network layer. RIP is the Fast Propagation of good news and the slow propagation of bad news. (2) OSPF (Open shortest Path First): the shortest Path First has priority. three key points are as follows: information is sent to the vro of the autonomous system by the flood method (no longer sent to the Information sender ); the sent information is the link status of all routers adjacent to the current vro, but this is only part of the information that the vro knows. Only when the link status changes, the router sends this information to all routers in flood mode. OSPF uses IP packet transmission directly, so OSPF is at the network layer. 2) EGP: External gateway protocol. If the origin site and destination site are in different autonomous systems, when the datagram is transmitted to the boundary of an autonomous system, you need to use one protocol to transmit route selection information to another autonomous system, such as BGP. 58. Does the adaptive Nic only flash at a red light or the green light? Is this normal? Normal. The adaptive Nic red light indicates connection/operation, that is, when the connection is made, the red light is always on, and when data is transmitted, the green light indicates full duplex, that is, the full duplex status is on, and the half duplex status is off. If a half-duplex network device (such as the HUB) is connected to an adaptive network, because the network adapter is an adaptive network adapter, it will work in the half-duplex status, so it is normal that the green light is not bright. Supplement: The NIC traffic light is the indicator light for Nic operation. The red light indicates that data is being sent or received, and the green light indicates that the network connection is normal. Therefore, it is normal that the green light is always on, because the green light is always on the network. When there is data transmission, the red light will flash. 59. ping failure after two laptops are connected. What problems do you think may exist? (1) first consider whether it is a network problem (hardware connectivity) (2) LAN Settings, Computer Interconnection needs to be set. Check whether necessary network protocols are installed. The most important thing is whether the IP address is set correctly. (Software protocol installation, IP settings) (3) Incorrect NIC Driver Installation (Driver Installation) (4) firewall settings (firewall) (5) is there any software that blocks ping packet (rogue software) 60 and other protocols that are compatible with the IP protocol? ARP: IP Address Resolution Protocol MAC address RARP: Reverse Address Resolution Protocol ICMP: Internet Control Packet protocol IGMP: Internet Group Management Protocol relationship: 61, IP address classification? IPv4 addresses have A total of 32bit network numbers. Network range: host number A class 8bit first fixed 00 -- 12724 bitB Class 16 bit first two fixed 10128.0 -- 191.25516 bitC class 24bit first three fixed as 110192.0.0 -- 223.255.2558bitD class the first four digits are fixed at 1110, the first five digits of the Class-D multicast address Class E are fixed to 11110, and the back ends are reserved for future use. Generally, all 0 or all 1 addresses are not used, which has special meanings, when the host address is full 1, the broadcast address is displayed. If the host address is full 0, the network address is displayed. At the same time, 127.0.0.1 indicates a loop. ping this IP address to test whether the local TCP/IP protocol is successfully installed. 62. RARP? The Reverse Address Resolution Protocol maps hardware addresses to IP addresses. It is mainly used for diskless workstations because the IP addresses configured for diskless workstations cannot be saved. Workflow: Configure an RARP server in the network, which stores the ing between the IP address and the MAC address. When the diskless workstation is started, it encapsulates an RARP packet, it contains its MAC address and then broadcasts it to the network. When the server receives the request packet, it finds the IP address of the corresponding MAC address and loads it into the response packet and sends it back to the requester. Due to the need to broadcast request packets, RARP can only be used in networks with broadcast capabilities. 63. Divide subnets? In general, the classification method is similar to that of only the network number and host number. This is done by the Network allocated to the network number, a subnet is formed by borrowing a bit from the host number. When a subnet is involved, a subnet mask is required, the network number of an IP address involving the subnet is equal to the sum (&) operation between the IP address and the subnet mask. 64. IPv6? BITs are used, and the first fixed part is 40 bytes. 65. What is the difference between transport layer protocol and network layer protocol? What is the network layer protocol responsible for the provision of logical communication between hosts? What is the transport layer protocol responsible for the provision of logical communication between processes? TCP: the unit of transmission is TCP packet segment UDP. The unit of transmission is user datagram. Its port is used to identify which application is using this protocol. 67. What interfaces are used for access networks? Generally, interfaces such as E1, V.24, and V.35 are used.