An article on HTTP and TCP protocols

[[397802]]

This article is reprinted from the WeChat public account "Learn Front-end in Three Minutes", author sisterAn. Please contact the WeChat public account "Learn Front-end in Three Minutes" to reprint this article.

introduction

This article starts from the OSI network layering (7 layers) to explore the relationship between TCP and HTTP, including the following parts:

• OSI network layering (7 layers)
• TCP protocol (three-way handshake, four-way wave)
• HTTP
• Differences and connections between TCP and HTTP

OSI network layering (7 layers)

Open Systems Interconncection

1 Physical layer -> 2 Data link layer -> 3 Network layer (ip) -> 4 Transport layer (tcp) -> 5 Session layer --> 6 Presentation layer --> 7 Application layer (http)

HTTP is an application layer protocol, and TCP is a transport layer protocol. Next, we will introduce them in detail one by one.

TCP

TCP and UDP are both transport layer protocols:

• User Datagram Protocol UDP (User Datagram Protocol):
  • No connection;
  • Best efforts delivery;
  • Message oriented;
  • No congestion control;
  • Support one-to-one, one-to-many, many-to-one, and many-to-many interactive communications;
  • The header overhead is small (only four fields: source port, destination port, length, and checksum).
• Transmission Control Protocol TCP (Transmission Control Protocol):
  • Connection-oriented;
  • Each TCP connection can only be point-to-point (one-to-one);
  • Provide reliable delivery services;
  • Provide full-duplex communication;
  • Byte stream oriented.

In addition, UDP is a message-oriented transmission method. The application layer sends the message of the same length as the message given to UDP, that is, one message is sent at a time. Therefore, the application must choose the appropriate message size.

The interaction between the application and TCP is one data block (of varying sizes) at a time, but TCP sees the application as a series of unstructured byte streams. TCP has a buffer, and when the data block sent by the program is too long, TCP can divide it into shorter pieces before sending it.

When TCP protocol is used for network communication, a connection must be established between the client and the server before the actual read and write operation. When the read and write operation is completed, the connection can be released when both parties no longer need it. The establishment of the connection relies on the "three-way handshake", while the release requires the "four-way handshake", so the establishment of each connection requires resource consumption and time consumption.

TCP connection process (three-way handshake)

First Handshake

The client sends a connection request segment to the server. This segment contains its own data communication initial sequence number. After the request is sent, the client enters the SYN-SENT state.

Second handshake

After the server receives the connection request segment, if it agrees to connect, it will send a response, which will also include its own data communication initial sequence number. After sending, it will enter the SYN-RECEIVED state.

The third handshake

When the client receives the connection agreement, it also sends a confirmation message to the server. After the client sends this message segment, it enters the ESTABLISHED state. After the server receives this response, it also enters the ESTABLISHED state. At this time, the connection is successfully established.

Why do we need three handshakes? Two is not enough?

Hello, hello, this is A, can you hear me? B: Yes, yes, I can hear you, this is B, can you hear me? A: (I heard you, I don’t want to talk to you) B: Hello, hello, hello? Can you hear me? Fuck, the other party is dead, I’m dead. .

If there are only 2 times, B has no idea whether A received the message he sent.

TCP disconnects the connection (four times wave)

First wave

If client A thinks that the data transmission is completed, it needs to send a connection release request to server B.

Second wave

After receiving the connection release request, B will tell the application layer to release the TCP connection. Then it will send an ACK packet and enter the CLOSE_WAIT state, which means that the connection from A to B has been released and will no longer receive data sent by A. However, because the TCP connection is bidirectional, B can still send data to A.

The third wave

If B has unsent data at this time, it will continue to send it. After completion, it will send a connection release request to A, and then B will enter the LAST-ACK state.

PS: Through the delayed confirmation technology (usually with a time limit, otherwise the other party will mistakenly think that retransmission is required), the second and third handshakes can be combined to delay the sending of the ACK packet.

The fourth wave

After receiving the release request, A sends a confirmation response to B, and then A enters the TIME-WAIT state. This state will last for 2MSL (the longest segment life, which refers to the time a segment survives in the network. If it times out, it will be discarded). If there is no retransmission request from B during this period, it will enter the CLOSED state. When B receives the confirmation response, it also enters the CLOSED state.

HTTP

HTTP is an application layer protocol built on TCP, Hypertext Transfer Protocol.

The most notable feature of HTTP connection is that each request sent by the client requires the server to send back a response, and after the request is completed, the connection will be released actively. The process from establishing a connection to closing the connection is called "one connection".

http1.0: Each request from the client requires a separate connection to be established, and the connection is automatically released after the request is processed.

http1.1: Multiple requests can be processed in one connection, and multiple requests can be overlapped. There is no need to wait for a request to end before sending a new request.

http2.0: supports multiplexing, one TCP can transmit multiple http requests at the same time, and the header data is also compressed

http3.0: QUIC is used to open multiple TCP connections. In the event of packet loss, only the TCP connection with the packet loss will wait for retransmission, and the remaining TCP connections can still transmit data normally.

HTTP Features

• Stateless: The protocol has no state storage for the client and no "memory" ability for transaction processing. For example, accessing a website requires repeated login operations.
• Connectionless: Before HTTP/1.1, due to its stateless nature, each request required four handshakes through the TCP three-way handshake to reestablish a connection with the server. For example, if a client requests the same resource multiple times in a short period of time, the server cannot distinguish whether it has already responded to the user's request, so it needs to respond to the request again each time, which consumes unnecessary time and traffic.
• Request and response based: The basic feature is that the client initiates the request and the server responds.
• Simple, fast and flexible.
• The communication uses plain text, and the requests and responses do not confirm the communicating parties, and the integrity of the data cannot be protected.

Method

• The GET method requests a representation of a specified resource. Requests using GET should only be used to retrieve data.
• The HEAD method requests a response that is the same as the response to a GET request, returning only the request header, without a response body, and is mostly initiated by JavaScript
• The POST method is used to submit an entity to a specified resource, typically resulting in a change of state or side effects on the server.
• The PUT method replaces any current representation of the target resource with the request payload.
• The DELETE method deletes the specified resource.
• The CONNECT method establishes a tunnel to the server identified by the target resource and is commonly used for HTTPS and WebSocket.
• OPTIONS method, pre-check, is used to describe the communication options of the target resource. Through this request, we know whether the server allows cross-domain requests.
• The TRACE method performs a message loopback test along the path to the target resource, which is not supported by most online services.
• The PATCH method is used to apply partial modifications to a resource.

Differences between HTTP and TCP

The TCP protocol corresponds to the transport layer, while the HTTP protocol corresponds to the application layer. In essence, the two are not comparable:

• HTTP corresponds to the application layer, and TCP protocol corresponds to the transport layer
• The HTTP protocol is built on top of the TCP protocol. When HTTP initiates a request, it establishes a channel to connect to the server through the TCP protocol. After the request is completed, the TCP connection is immediately disconnected.
• HTTP is a stateless short connection, while TCP is a stateful long connection
• TCP is a transport layer protocol that defines the specifications of data transmission and connection methods, while HTTP is an application layer protocol that defines the specifications of the content of the transmitted data.

Note: Since HTTP/1.1, Keep-Alive has been enabled by default. Simply put, when a web page is opened, the TCP connection used to transmit HTTP data between the client and the server will not be closed. If the client accesses the web page on the server again, it will continue to use the established connection. Keep-Alive will not maintain the connection permanently. It has a hold time, which can be set in different server software (such as Apache).

Source: https://github.com/Advanced-Frontend/Daily-Interview-Question