Shulou(Shulou.com)06/02 Report--

6.1introduction to HTTP

Web, servers, and related Web applications all communicate with each other through HTTP. HTTP (HyperText Transfer Protocol, Hypertext transfer Protocol) is a common language used in modern global Internet. Web content is stored on Web servers, and Web servers use the HTTP protocol, so they are often referred to as HTTP servers. These HTTP servers store data from the Internet and provide data if the HTTP client requests it. The client sends a HTTP request to the server, and the server sends back the requested data in the HTTP response.

Resources and types of resources

The Web server is the host of Web resources, and Web resources are the source of Web content. The simplest Web resource is the static file in the Web server file system. These files can contain anything: text files, HTML files, word files, AVI movie files, and so on. But resources are not necessarily static files, resources can also be software programs that generate content according to needs, and these programs can generate different content according to the identity of the user and the requested information.

There are thousands of different data types on the Internet, and HTTP labels each object to be transmitted through Web with a data format of MIME type. The Web server appends the MIME (Multipurpose Internet Mail Extension, Multipurpose Internet Mail extension) type to all HTTP object data. When a Web browser retrieves an object from the server, it looks at the relevant MIME type to see if it knows what to do with the object. Most browsers can handle hundreds of common object types.

The MIME type is a text tag that represents a major object type and a specific subtype, separated by a slash.

Text documents in HTML format are marked by the text/html type

Ordinary ASCII text documents are marked by the text/plain type

The JPEG version picture is of image/jpeg type.

The GIF version picture is of image/gif type.

...

URI and URL

Each Web server resource has a name so that the client can find the resource it needs. The server resource name is called the uniform Resource Identifier (Uniform Resource Indentifier,URI). URI, like the postal address on the Internet, uniquely identifies and locates information resources around the world.

For example: http://www.joes-hardware.com/specials/saw-blade.gif

Uniform Resource Locator (Uniform Resource Locator,URL) is the most common form of resource identifier, and URL describes the specific location of a resource on a particular server. They can specify how to obtain resources from a precise, fixed location. The figure shows how URL accurately describes a voluntary location and how to access it.

Business

The following figure shows how the client transacts with the Web server and its resources through HTTP. An HTTP transaction consists of a request (from the client to the server) and a response (the server sends back to the client). This communication takes place through the formatted data blocks of the HTTP message.

HTTP supports several different request commands, which are called HTTP methods (HTTP mothod). Each HTTP request message contains a method. These methods tell the server what action to perform, and common HTTP methods are:

When GET requests to obtain a resource, it requires the server to send HEAD similar to GET, but it does not need the service to send resources, but only sends back the response header PUT, which is opposite to GET, and writes a document to the server. For example, the publishing system DELETE requests to delete the resource OPTIONS pointed to by URL.

Probe the request method TRACE supported by the server for a resource

POS such as firewalls, agents, or gateways to track requests to go through

Support for HTML form submission, where there is data filled in by the user, which is sent to the server and stored in a location (such as a send handler)

Each HTTP response message is returned with a status code. A status code is a three-digit code that tells the client whether the request was successful or whether other actions need to be taken. Common status codes are:

1xx pure informational status code 2xx

Information for the "success" class (200,201,202) 3xx

Redirect status code

301:Moved Permanently, specify the current location of the resource with the header "Location: URL" in the response message

302:Found, using the header "Location: URL" in the response message to specify the temporary resource location

304:Not Modified, used in conditional requests; (used in caching)

4xx

Information about the client error class

403:Forbidden, request rejected by server

404:Not Found, the server cannot find the requested URL

405:Method Not Allowed, it is not allowed to request the corresponding URL using this method

5xx

Server-side error class information

500:Internal Server Error, server internal error

502:Bad Gateway, the proxy server received a pseudo response from the upstream

503:Service Unavailable, the server cannot provide service at this time, but may be available in the future

504:Gateway Timeout, when the gateway proxy requests the backend service, the backend service does not respond in time.

Message

HTTP messages are made up of simple strings on a line. Http messages are plain text, not binary code, so they can be easily read and written. The following figure shows the HTTP message used by a simple transaction.

The HTTTP message is mainly divided into three parts: the starting line, the header field and the body.

Request message syntax:

(name:value)

(must be a blank line)

Message body

Response message syntax:

(must be a blank line)

Request method, the actions you want the server to perform, such as GET, HEAD, POST, etc.

The requested resource can be either a relative path or a complete URL

Protocol version, format HTTP/., such as http/1.0

: HTTP header

Status code

Reason phrase, numeric status code readable information

: body part

HTTTP protocol version

Http/0.9: plain text only (hyperlinks), ASCII

Http/1.0: supports multimedia data processing. Staying connected is supported.

Http/1.1: persistent connections are supported. Finer cache control

HTTP-NG, also known as HTTP/2.0, focuses on significant performance optimizations and a more powerful remote execution framework for service logic, but has no plans to replace http/1.1.

HTTP header

The HTTP header works with the method to determine what the client and server can do. The first part can usually be divided into five types:

General header: both request message and response message can be used

HTTP/1.0 introduces headers that allow HTTP applications to cache local copies of objects so that they do not always need to be fetched directly from the server segment.

Request header: meaningful only in the request message

The informational header of the request:

First part of Accept:

Conditional request header:

Security request header:

Proxy request header:

Response header: meaningful only in the request message

Informational header of response

First part of negotiation

Security response header

Entity header: used to specify entity attributes

The informational header of an entity

Content header

Entity cache header

Extension header: non-standard header, which may be created by program developers, such as X-Forward-For

Implementation Software of HTTTP

Client:

GUI:IE 、 Firefox 、 Safari 、 chrome 、 Opera

CLI:elinks, curl

Server:

Server: httpd (ASF Foundation), nginx, lighttpd

App Server: IIS 、 tomcat 、 jetty 、 resin

6.2. the architectural way in which the Web server handles connection requests

1. Single-thread web server (Single-threaded web servers)

In this architecture, the web server processes one request at a time, and then reads and processes the next request. During the processing of a request, all other requests will be ignored, so there will be serious must-be problems in scenarios with more concurrent requests.

2. Multi-process / multi-thread web server

In this architecture, the web server generates multiple processes or threads to process multiple user requests in parallel, and the processes or threads can be generated on demand or in advance. Some web server applications generate a separate process or thread for each user request to respond, but once the number of concurrent requests reaches thousands, multiple processes or threads running at the same time will consume a lot of system resources.

3. Icano multiplexed web server

In order to support more concurrent user requests, more and more web servers are adopting a multi-reuse architecture-synchronously monitoring the activity status of all connection requests, and performing a series of specific operations for a connection when the state of a connection changes (such as data preparation or an error occurs) After the operation is complete, the connection will return to a temporary steady state and return to the list of open connections until the next state change. Because of its multiplexing nature, processes or threads will not be occupied by idle connections, so it can provide an efficient working mode.

4. Multiplexing multithreaded web server

The web server architecture, which combines the functions of multi-process and multiplexing, avoids making one process serve too many user requests, and can make full use of the computing power provided by multiple CPU hosts.

The process of a web request response

1. Establish a connection-accept or reject client connection requests

2. Receive the request-read the HTTP request message through the network

3. Process the request-parse the request message and take corresponding actions

4. Access resources-- access related resources in the request message

5. Build the response-use the correct header to generate the HTTP response message

6. Send response-send the generated response message to the client

7. Log-- when completed HTTP transactions are recorded in the log file

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.