Shulou(Shulou.com)06/01 Report--

How to parse the micro-service architecture SpringCloud, aiming at this problem, this article introduces the corresponding analysis and solution in detail, hoping to help more partners who want to solve this problem to find a more simple and feasible method.

To work hard is not to move anyone, nor to show it to some people, but to have the ability to jump out of the circle you hate and have the right to choose to live your life in the way you like.

Micro-service architecture (generally using AP architecture)

A: service availability

P: partition fault tolerance

C: strong consistency of data

Domain-driven design: through code and data analysis to find a reasonable syncopation point, and through data analysis to determine the service division boundary and division granularity.

Microservice core: Docker is the container center, DevOps is a deployment tool or concept (DevOps is the merger of Development and Operations)

Three difficult problems in micro-service architecture: propagation of service failures, division of services, and distributed transactions

Avalanche effect: service unavailability causes the system to be paralyzed (resolved by fuses)

Microservices have the following functions:

Registration and Discovery of Services

Load balancing of services

Fault tolerance of services

Service gateway

Service configuration center

Service link tracking

Compared with other service frameworks

Dubbo: an excellent framework for service governance and service invocation

Kubernets: achieve all the micro-service functions through container choreography, more like a platform

Maven configuration

.settings.xml local warehouse directory, remote download server (foreign server, Aliyun mirror library, build your own mirror library-Sonatype Nexus Repository Manager)

Pom.xml files: groupid, artifactid, version, parent, properties, dependencies, build

Maven commands: mvn clean, mvn package, mvn compile, mvn install

Spring boot:

Three major features: automatic configuration, start-up dependence, and Actuator monitoring of health

/ / @ RestController = @ Controller+@ResponseBody

@ RestController

Public class HelloController {

@ RequestMapping ("/ hello")

Public String index () {

Return "hello spring boot"

}

}

Springboot profile

Application.yml configuration files are loaded into configuration classes, and custom configuration files are loaded into classes

Configuration files for multiple environments: application- {profile} .properties test, dev, prod

Need to add spring.profiles.active = dev configuration to application.yml

Actuator monitoring

Configure the application.yml property to support management.port and management.security.enabled

Http://localhost:9091/health to view the running status of the program

/ beans View the bean of the running program

Springboot integrates jpa and redis

Springboot integrates swagger2, builds RestfulAPI online documentation, @ API @ ApiOperation (value= "user list", notes= "user list")

Service Registration and Discovery Eureka

Register Service: service registry

Provider Service service provider application server

Consumer Service service consumer client sever

Some concepts:

Register: service registration-ip, duank, status, home page access address

Renew: service renewal client by default sends a heartbeat every 30 seconds for service renewal

Fetch Registries: get service registration list information. Client is updated once by default at 30.

Cancel: service offline

Eviction: service is removed. By default, client does not send renewal service for 90 seconds, and sever removes it from the service list.

Build a highly available Eureka Server cluster: application.yml is in multiple profile format, divided into host and port

/ / * prevent yourself from registering yourself *

Eureka:

Client:

Register-with-eureka: false

Fetch-registry: false

/ / *

Load balancing Ribbon

There are two ways to implement load balancing: one is an independent process unit, which is forwarded to different execution units through load balancing, eg:nginx, and the other is to encapsulate the load balancing logic in code to the client of the service consumer, eg:Ribbon.

RestTemplate is combined with Ribbon, and the @ LoadBalanced annotation is added to resttemplate bean to enable load balancing.

2 eureka-client service consumers can access it according to the Eureka service name

The declaration call to Feign

The goal of Feign is to make the java Http client call process simple.

FeignClient (value= "eureka-client", configuration=FeignConfig.class)

Fuse Hystrix

Have the function of self-repair

@ EnableHystrix

@ HystrixCommand (fallbackMethod= "hiError")

Hystrix Dashboard monitoring

Routing Gateway Spring Cloud zuul

Unified aggregation of all api interfaces

Can do user identity and authority authentication

Real-time monitoring, log book

Traffic monitoring

Zuul includes four types of filters:

PRE filter: executed before the request to a specific service, and can be used for authentication and parameter verification

ROUTING filter: executed when a request is routed to a specific microservice instance

POST filter: executed after the request has been routed to the microservice and can be used to collect log information, metrics, and responses

ERROR filters: executing when errors occur in other filters

Route distribution, support version number

Fuses can be configured on the route

Zuul is easy to implement load balancing, intelligent routing and fuses, and its scale-out capability is very good.

Common implementation methods: 1. Different channels use different routes (app, web, etc.)

2. Combined with nginx, the outermost exposure is that Nginx master-slave double hot backup keepalive,Nginx forwards the request to the Zuul cluster through some policy, and Zuul finally distributes the request to the specific service.

Configure Central Spring Cloud Config

@ EnableConfigServer

1. Read the configuration file locally

2. Read the configuration file from the git server

Using Spring Cloud Bus to refresh the configuration, optional message broker components include RabbitMQ, AMQP, and Kafka.

To change the configuration file, a function of Bus makes this process simple. You only need to send a Post request to a micro-service instance and notify other micro-service instances through the message component to pull the configuration file again.

Config-client, implemented through RabbitMQ, needs to build a MQ service, add @ RefreshScope to the client main class, and send a request through Postman or other tools, http://ip:port/bus/refresh

Service Link tracking SpringCloud Sleuth

Use Zipkin to provide UI presentation, use Http or RabbitMQ to transfer link data, use mysql, ElasticSearch (suitable for large data access, you can use Kibana to display link data), and Cassandra database to store link data.

Micro service monitoring Spring Boot Admin

Monitor Spring Cloud micro-service, aggregate monitor fuse status in micro-service system, integrate Security security login interface

Spring Boot Security security components

System security also needs to consider the transport layer and system layer, eg:https, server fire window.

Peer excellent security framework Apache Shiro, generally used for single applications.

User information can be stored in memory or in the database.

Spring Cloud OAuth3 protection microservice system

OAuth3 is a standard authorization protocol

OAuth3 authentication process, obtaining Token process, using token to request authentication, authentication by taking user permissions

The drawback of the framework is that it has to remotely request auth-service to verify the validity of token every time.

JWT (JSON WEB TOKEN) is an open standard with very small data sizes.

Application scenarios: authentication, information exchange, you can use JWT to achieve single sign-on

A large number of api access logs can be processed using the ELK component

This is the end of the answer to the question on how to analyze the micro-service architecture SpringCloud. I hope the above content can be of some help to you. If you still have a lot of doubts to solve, you can follow the industry information channel for more related knowledge.

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.