Interviewer: What process will be executed after entering the URL?

After entering the URL in the browser, it will perform the following processes:

Perform DNS domain name resolution;

Encapsulate HTTP request data packets;

Encapsulate TCP request data packets;

Establish a TCP connection (3-way handshake);

Parameters are passed from the client to the server;

After receiving the client parameters, the server performs corresponding business processing, encapsulates the results into HTTP packets, and returns them to the client.

The interaction between the server and the client is completed, and the TCP connection is disconnected (4 waves);

The browser uses its own execution engine to render and display the final result to the user.

1. DNS domain name resolution

Location in the network relies on IP for identity location, so the first step of URL access is to get the IP address of the server. To get the IP address of the server, you need to use DNS (Domain Name System) domain name resolution, which is to find the corresponding IP address through the URL.

PS: Why not directly access the IP address to request the server? Because IP addresses are long and inconvenient to remember, while URL addresses are much easier to remember, so URLs are used to replace IP addresses. URLs are like aliases of IP addresses, and can be used to locate the corresponding IP addresses.

The general process of DNS domain name resolution is as follows:

1. First check the DNS cache in the browser. If there is a corresponding record in the browser, it will be used directly and the resolution will be completed;
2. If the browser does not have a cache, it will query the cache of the operating system. If a record is found, it can directly return the IP address to complete the resolution;
3. If the operating system does not have a DNS cache, it will check the local host file. In Windows operating systems, the host file is generally located in "C:\Windows\System32\drivers\etc\hosts". If the host file has a record, it will be used directly;
4. If there is no corresponding record in the local host file, the local DNS server will be requested. The local DNS server is generally provided by local network service providers such as China Mobile and China Telecom. Usually it can be automatically assigned through DHCP, of course you can also configure it manually. Currently, the most commonly used public DNS is 8.8.8.8 provided by Google and the domestic public DNS is 114.114.114.114.
5. If the local DNS server does not have the corresponding record, it will query the root domain name server. There are currently 13 groups of root domain name servers in the world (this does not refer to 13 servers, but 13 IP addresses, numbered by letters AM). In order to complete the resolution requests of all domain names in the world more efficiently, the root domain name server itself will not directly resolve the domain name, but will assign different resolution requests to other servers below to complete. The following is a tree structure diagram of the DNS domain name system:

2. Encapsulate HTTP request data packets

An HTTP request object contains 4 parts:

1. Request Line
2. Request Header
3. Blank lines
4. Request body

Its basic format is as follows:

After obtaining the server IP, the local request will be encapsulated into an HTTP data packet, as shown in the figure above.

3. Encapsulate TCP request data packet

The underlying HTTP layer is implemented based on the TCP/IP protocol, so during the underlying data transmission, the HTTP request packet will be further encapsulated into a TCP data packet.

4. Establish a TCP connection (3-way handshake)

The basis of HTTP communication is TCP connection. TCP connection requires three handshakes. The three handshakes are to verify the sending and receiving capabilities of the client and the sending and receiving capabilities of the server. Just like a phone call, a normal call starts like this:

1. _Me_: Hello, can you hear me?
2. _Other party_: Yes, can you hear me? (This proves the other party's receiving ability and my sending ability)
3. _Me_: I can hear it too, let's talk business. (Proving the other party's sending ability and my receiving ability)

After the above three handshakes, the sending and receiving capabilities of the client and the sending and receiving capabilities of the server can be proved, so that communication can officially begin.

5. The server obtains the HTTP request parameters

After the data is transmitted to the server program through TCP, the TCP data packet will be converted into HTTP data packet (all this is thanks to the TCP/IP protocol), so that the server can get the request data sent by the client.

6. The server performs business processing and returns data

After the server receives the request parameters from the client, it will perform the corresponding business processing. After the processing is completed, it will return the processing results to the client. The return process is similar to the sending process. The result is first encapsulated into an HTTP data packet. The HTTP data packet can be divided into the following 4 parts:

1. Status Line
2. Response Header
3. Blank lines
4. Response body

Its basic format is as follows:

The status line is used to describe the return status of the server. It consists of 3 parts:

1. HTTP version number, such as HTTP/1.1;
2. Status code, such as 200;
3. Status description information, such as OK;

The common status codes are as follows:

• 200: return success;
• 301: Permanent redirect;
• 302: Temporary redirect;
• 404: Page not found;
• 500: Server program error.

The response body is all the data returned to the client.

7. Disconnect TCP connection (4 waves)

After one request and one response, the "communication" between the client and the server ends, and the TCP disconnection process can be executed, which requires 4 waves:

• _Client_: Let’s break up;
• _Server_: OK, let me get ready.
• _Server_: I'm ready, let's break up.
• Client: OK.

After the above process, the TCP connection is disconnected.

8. The browser renders and displays the results

After TCP interaction, the client also receives the data returned by the server, and then uses the browser's own execution engine to display the final result to the user, and the entire execution process is completed.

References & Acknowledgements

zhuanlan.zhihu.com/p/58108010​