Linux sends udp broadcast packets

Tcpudp testing tool (tcp debugging assistant) is an excellent network protocol debugging software. How to debug tcp? The tcpudp testing tool (tcp debugging assistant) is convenient for users. This tool can send and receive TCP and UDP data. Download it if you need it!

Software functions:

1. supports TCP protocol to send and receive data. Support server mode and client mode. Server mode is mainly used to test TCP reception. The client is mainly used for TCP active requests.

2. Support UDP and UDP broadcast mode to send and receive.

3. Support multiple connections and operate multiple network connections at the same time.

4. Support sending Chinese characters.

5. Support hexadecimal sending and receiving mode.

Is tcp directly connected or udp?

TCP is link-based (reliable understanding must be established before sending and receiving), while UDP is connectionless (source and terminal do not need to establish a link, so they can send whenever they want).

TCP connection needs to go through three-way handshake, disconnection needs to go through four handshakes, UDP does not, direct connection is sent, direct disconnection (regardless of the other party);

TCP is a streaming mode (oriented to byte stream), while UDP is a datagram mode (oriented to message) (meaning that TCP must have a pipeline to support its continuous transmission, while UDP is like hitting a plane, if there is a bullet, it will be stuck, and if there is no bullet, it will stop);

The header packet size of TCP is 20~24 bytes (with 4 spare bytes) followed by the actual data, and the header packet size of UDP is 8 bytes followed by the actual data;

UDP is more effective than TCP because of the different packet sizes.

TCP (reliable) can guarantee the correctness and order of data, while UDP (unreliable) may lose packets and cannot guarantee the order.

TCP uses flow control and congestion control, while UDP does not.

TCP needs to establish a socket for each client, while UPD only uses one socket for communication.

TCP has no limit on the packet size, so there is no need to segment the packet separately. The underlying program of TCP has done segmentation for us. UDP is based on datagrams, which are limited in size and need to be cut by themselves. If you don't cut it, it will be too big and lose the packet.

TCP itself has a checksum retransmission mechanism, while UDP does not.

TCP only supports unicast transmission, while UDP supports unicast multicast and broadcast.

TCP throughput is regulated by congestion control algorithm, while UDP is only limited by data generation rate of application software, transmission bandwidth and performance of source host.

The similarities between TCP and UDP are:

All are long links;

Both require connection and disconnection event support;

Both send and receive queues need to be cached;

Need to define the package format (that is, the language of communication between client and server, such as JSON);

Need to encrypt and verify the data (otherwise it is easy to be inserted);

The data size after the first packet is 2 16 (65535 bytes, more than 60 KB);

It is easy to lose packets when transmitting more than 50 kb each time. TCP cuts automatically, while UDP needs to be cut manually.

The data packet cannot be too small, too small means that multiple data packets need to be sent at the same time, which is easy to lose packets;

In the OSI seven-layer model, both belong to the transport layer (transport layer: responsible for grouping messages and encapsulating them into TCP or UDP protocol format);

What is the principle of udp broadcasting?

Principle: udp user datagram protocol (RFC768)

User Datagram Protocol (UDP) is a connectionless transport layer protocol in OSI reference model, which provides simple and unreliable transaction-oriented information transmission service. UDP protocol is basically the interface between IP protocol and upper layer protocol. UDP protocol is suitable for multiple applications running on the same device.

Since most network applications run on the same machine, the computer must be able to ensure that the software program on the destination machine can get the data packet from the source machine, and the source computer can receive the correct reply.

What is the meaning of Internet technical terms?

Noun interpretation

Address: the unique identification of a computer or node on the Internet, which can be an IP address expressed by numbers or an address expressed by a domain name. ADSL: Asymmetric digital subscriber line, which uses two pairs of telephone lines to realize high-speed digital connection.

ASCII: American standard code for information interchange code, with 96 displayable uppercase and lowercase letters and 32 non-display control characters.

Bridge: A bridge is a device that connects two network node components that use the same protocol, and it is also a device that realizes data exchange between different networks in a local area network.

Computer network: a system in which computers, terminals and other devices with independent functions are connected through communication lines, communicate in a certain way and share resources.

Centralized network: it is a network with a star or tree topology, in which all information passes through the central node switch and all links originate from the central node switch.

Decentralized network: The concentrator and multiplexer have a little switching function, which is a mixture of star network and grid network.

Distributed network: Its characteristic is that any node is directly connected with at least two other nodes, which has high reliability.

WAN: Its range of action is usually tens to thousands of kilometers.

Local area network: its scope of action is generally a building or a small unit, about one kilometer.

ISO has seven levels of architecture: physical layer, data link layer, network layer, transport layer, session layer, presentation layer and application layer. TCP/IP protocol is a network architecture and protocol standard developed by the United States for ARPANET in the mid-1970s.

Internet: A physical collection of routers/switching devices/computers and lines based on a set of * * * identical protocols or a set of * * * shared resources, which can even be regarded as a method of interconnection and information exchange between networks.

Domain name: the complete expression of Internet nodes, including host name, subdomain name and domain name, all separated by dots. Every host on the Internet has a unique IP address.

E-mail: E-mail is the most widely used service and the most important and basic application of the Internet. It can send and receive information from various media, such as words, images and sounds, and can be sent to multiple recipients at the same time or forwarded to a third party. It's slower than real-time fax, but much cheaper. E-mail on the Internet is a very convenient communication tool, which has been widely used from its early application in academic discussions to now, especially in business communication, and has great development potential.

Telnet: Telnet is an early service provided on the Internet. Users make their computers temporarily become the terminals of remote computers through Telnet commands, and directly call the resources and services of remote computers. Using remote login, users can use all the resources open to the outside world on remote computers in real time, query databases, retrieve materials, or use remote computing to complete tasks that only supercomputers can complete. In addition, many Internet services are realized through Telnet access.

File transfer (FTP): FTP(FileTransferProtocol) is one of the traditional Internet services. FTP enables users to transfer files between two networked computers, which is the most important method to transfer files on the Internet. Using anonymous FTP, you can get rich Internet resources for free. In addition, FTP also provides session control functions such as login, directory query and file operation.

Newsgroups: Newsgroups are user groups organized by users with the same interests on the Internet. News refers to some special discussion groups of news groups. Newsgroup has more than 6000 discussion groups. Each group revolves around a theme. It involves computer, biology, mathematics, philosophy, politics, economy, social problems, jokes, science fiction sketches, astronomy and geography, fashion, tourism and so on.

Finger:Finger is a utility program used to query the user's situation in UNIX system. UNIX system keeps detailed information of each user, including e-mail address, account number, real name in real life, login time, unread letters, the last time to read e-mail and messages when going out. When you use the finger command to query, the system will display the above information one by one on your terminal or computer.

Gopher:Gopher is a menu-driven information query tool provided by the Internet, which adopts client/server mode. There are thousands of Gopher servers on the Internet. They organized the information resources of the Internet into a single form of database called Gopher space. Gopher is different from general information query tools. It uses keywords as an index, and users can easily connect from one Internet host to another to find the information they need.

Wais: Wide-area information service Wais is a fast information query service tool provided by the Internet. Wais organizes relevant information into a database, which can be searched by keywords and provided to the public.

The World Wide Web is also an information service system based on the Internet. But it is more flexible and easier to provide information services than Gopher. Based on the so-called hypertext technology, WWW uses file-oriented browsing instead of the usual menu list. Can provide a certain format of text and graphics. Web links global information resources through keywords, so that information can not only be searched linearly, but also accessed cross.

Computer virus: By attaching to other programs, it automatically copies itself, or performs unnecessary operations, thus destroying normal programs and data, and even destroying the operating system.

Firewall: Set a barrier on the interface between the internal network and the external network to prevent unpredictable and potential access and damage to the internal network.

Host: a high-performance computer with network control and processing functions that provides services to users.

Communication subnet: It consists of some special communication processors (such as node switches) and communication links connecting these nodes.

Agreement: It is a pre-agreed rule that both parties must abide by.

10BaseT: technical name of twisted-pair Ethernet.

2-3Swap: refers to a wire whose one end is used for sending and the other end connected with it is used for receiving or vice versa. Numbers 2 and 3 refer to the transmit and receive pins of DB-25 connector.

2b+DService: ISDN service, because it contains two standard telephone connections and one data connection.

Three-way handshake Tcp three-way handshake: A technique used in TCP and other transport protocols to reliably start or end communication.

Three-wire circuit Three-wire circuit: A common wiring scheme for asynchronous serial connection between a pair of computers. The first wire is used to transmit data from one computer to another, the second wire is used to transmit data in the opposite direction, and the third wire is a common ground wire.

Four-wire circuit Four-wire circuit is a common wiring scheme for asynchronous serial connection between a pair of computers. One pair of wires is used to transmit data in one direction, and the other pair is used to transmit data in the opposite direction. Four-wire circuits are usually used for longer distances than three-wire circuits.

Seven-Layer Reference Model Seven-layer reference model: an early conceptual model promulgated by the International Organization for Standardization, which gives a series of protocols to match the communication services provided. The seven-layer protocol does not include the Internet protocol layer.

802.2:IEEE logical link control standard. See LLC and SNAp.

802.3:IEEE Ethernet standard.

802.5:IEEE Token Ring Standard.

Access delay: the waiting time of the network interface before accessing the network.

Acknowledgement response: A short return message informing the sender that the data has reached the expected destination.

Activedocument active document: WWW document is a computer program. After downloading the active document, the browser runs the program on the user's computer. The active document can constantly change its display. See dynamic documents, static documents and URLs.

Adaptive retransmission and adaptive repetition: the ability of an adaptive retransmission transmission protocol to constantly change the retransmission timer to adapt to various Internet delays. TCP is a well-known protocol that uses adaptive retransmission.

Addressmask: a 32-bit binary value, with each bit corresponding to the corresponding IP address of a network and subnet. The corresponding part of an address bit that is not covered by a mask, also known as a subnet mask.

Addressresolution Address resolution: Matching from address to address, usually from high-level address (such as IP address) to low-level address (such as large network address).

AnonymousFTP Anonymous File Transfer Protocol (FTP): Use the login name anonymons and a four-word guest to access the FTP accessor. Not all FTP servers allow anonymous FTP.

API (Application Programming Interface): A set of programs that computer programs can call to access specific services. A set of procedures used by programs to access network protocols are collectively called network API.

Applet: A computer program that constitutes an active World Wide Web document. Applets are written in programming languages such as Java.

AppleTalk: A set of network protocols developed and sold by Apple Computer Company.

ARP(AddressResolutionProtocol): A protocol used by computers to match IP addresses and hardware addresses. The computer calls the ARP broadcast request, and the target computer responds to the request.

ASCL (American standard code for information interchange code): a standard containing 128 unique characters, including uppercase letters, lowercase letters, numbers and punctuation marks.

Asn. 1 (abstract syntax notation. 1) abstract syntax notation 1: a standard for representing data. The SNMP protocol uses ASN. 1 to represent the object name.

Asynchronous: A characteristic of communication system, that is, the sender can send data without warning, and the receiver must be ready to receive data at any time. See synchronization.

ATM(AsynchronousTransferMode): A connection-oriented technology defined by ITU and ATM Association. The data sent by ATM at the lowest level is contained in a cell consisting of 48 octal data.

AUI(AttachmentUnitInterface) connection unit: used to connect thin cables in large networks. AUI connection appears in the connection between computer and Ethernet transceiver.

AWT(AbstractWindowToolkit) abstract window toolbox: A graphical process library, which uses Java language to manipulate windows through bitmap display. Then the designer extends AWT to AlternativeWindow Toolbox and AppletWidget Toolbox.

Bearerchannel)B channel: A term used by telephone companies to describe a channel configured to handle voice telephone lines. ISDN includes b channel service, see d channel.

Bandwidth: a measure of the capacity of a transmission system, and the bandwidth is in hertz.

Baseheader basic header: the header required at the beginning of IPv6 datagram.

Porter Porter: The number of signal changes per second. Each change can encode one or more bits of information.

Best-effort: The performance of any network system wants to transmit data optimally, but it cannot be guaranteed. Many networks do their best.

A technical protocol for dynamic address configuration. The computer randomly selects an address and broadcasts a message to confirm whether the address is in use, and uses the server management address interactively. See DHCP.

Binary power compensation: the scheme adopted by computers when Ethernet encounters conflicts, that is, each computer doubles the waiting time after each conflict.

BitspercondBinary bitsPersecond: the transmission rate of data to the network. The number of bits per second may be different from the baud rate of data, because a baud can be encoded with multiple bits.

BNC connector BNC connector: The connector type used by thin cable Ethernet.

Boot protocol: the protocol that the computer starts for the first time to obtain the information needed to configure the boot protocol software. BOOTP broadcasts requests using IP and UDP, and receives replies before IP is fully configured.

BRI (Basic Interface) Basic rate interface: two B channels and one data channel provided by ISDN services. BRI is suitable for small-scale operation. See PRI.

Bridge Bridge: A hardware device that connects two LAN segments and copies frames from one segment to the other. Many bridge hardware uses physical addresses to know which computer is associated with which network segment. In this way, the bridge can avoid copying frames unless absolutely necessary.

Broadcast: A form of transmission that sends a copy of a package to every computer on the network. See Clustering, Multicast and Unicast.

wide stretches of

What is the difference between TCP and UDP?

1, connection difference TCP is connection-oriented (for example, dial to establish a connection before making a phone call). UDP is connectionless, that is, there is no need to establish a connection before sending data.

2. Differences in security TCP provides reliable services, and the data transmitted through TCP connection is error-free, not lost, not duplicated, and arrives in order. UDP tries its best to deliver, that is, it does not guarantee reliable delivery.

3, the difference between transmission efficiency TCP transmission efficiency is relatively low. UDP has high transmission efficiency and is suitable for high-speed transmission and real-time communication or broadcast communication.

4. Difference in the number of connection objects TCP connections can only be point-to-point and one-to-one. UDP supports one-to-one, one-to-many, many-to-one and many-to-many interactive communication.