标签:conf alt iso maven插件 sim .com 运行时 icm voc
如何通过以下方式在Docker中运行Spring-Cloud MicroService拓扑:Docker端口映射到外部主机网络
Docker Links
Docker Bridge网络
Eureka和MicroServices可通过docker机器的静态IP地址访问。客户端只需获取eureka注册表并通过docker-machine的IP调用特定的MicroService,并将容器的端口映射到客户端的网络端口。
如果设置eureka.instance.prefer-ip-address = true,则使用Eureka的MicroService注册将使用eureka.instance.ip-address指定的IP地址(解析后默认为主机的IP)。
因此MicroService Spring Boot属性文件将是:
eureka.client.serviceUrl.defaultZone=http:<font><i>//${DOCKER_HOST_IP}:9761/eureka</i></font><font>
.
eureka.instance.preferIpAddress=<b>true</b>
eureka.instance.ip-address=${DOCKER_HOST_IP}
.
</font>
设置说明:
在Eureka上注册此MicroService,运行时间为: http://${DOCKER_HOST_IP}:9761 / eureka
MicroService将通过$ {DOCKER_HOST_IP}系统属性指定的IP暴露给docker的外部世界。我们将在这里注入docker机器的当前IP地址。
同样在 eureka server端配置:
eureka.instance.prefer-ip-address=<b>true</b>
eureka.instance.ip-address=${DOCKER_HOST_IP}
Eureka服务器实例将在$ {DOCKER_HOST_IP}上运行并可访问。
Docker maven集成
作为Docker Maven插件,我选择了 Spotify的Docker-Maven-Plugin 。功能齐全。
Eureka Dockerfile:
FROM java:8-jre
MAINTAINER Tomas Kloucek <tomas.kloucek@embedit.cz>
ADD ./demo-0.0.1-SNAPSHOT.war /app/
CMD [<font>"java"</font><font>, </font><font>"-Xmx512m"</font><font>, </font><font>"-jar"</font><font>, </font><font>"/app/demo-0.0.1-SNAPSHOT.war"</font><font>]
</font>
Eureka pom.xml (plugin configuration):
<plugin>
<groupId>com.spotify</groupId>
<artifactId>docker-maven-plugin</artifactId>
<version>0.4.11</version>
<configuration>
<imageName>${docker.image.prefix}/${project.artifactId}</imageName>
<dockerDirectory>src/docker</dockerDirectory>
<resources>
<resource>
<targetPath>/</targetPath>
<directory>${project.build.directory}</directory>
<include>${project.build.finalName}.war</include>
</resource>
</resources>
</configuration>
</plugin>
构建Docker镜像并在容器中运行它们:
(in the spring-microservice-registry folder)
mvn clean install docker:build
docker images (verify your image was built)
docker run --env DOCKER_HOST_IP=$(boot2docker ip) -it --name eureka -p 9761:9761 registry/demo
访问 http://DOCKER_HOST_IP:9761/ 确认eureka服务器正在运行
微服务客户端启动:
(in the spring-microservice-service folder)
mvn clean install -DskipTests docker:build
docker images (verify your image was built)
docker run --env DOCKER_HOST_IP=$(boot2docker ip) -it --name service -p 8080:8080 service/demo
docker ps (verify that service container is running)
从外部调用MicroService
外部客户只需要知道eureka服务器的地址
application.properties:
eureka.client.serviceUrl.defaultZone=http:<font><i>//${DOCKER_HOST_IP}:9761/eureka</i></font><font>
</font>
测试在Docker中启动的微服务:
(in the spring-microservice-client folder) mvn clean install java -jar -DDOCKER_HOST_IP=$(boot2docker ip) target/demo-0.0.1-SNAPSHOT.war
2.Docker Links
eureka服务器application.properties:
eureka.instance.prefer-ip-address = false
eureka.instance.hostname = eureka
通过此设置,Eureka服务器注册表将可以在地址 http://eureka:<port>/eureka 下载给客户端,
在Docker中启动Eureka注册表:
(in the spring-microservice-registry folder)
mvn clean install docker:build
docker images (verify your image was built)
docker run -it -p 9761:9761 --name eureka registry/demo
我只使用端口映射,因为我想使用eureka网页。重要的是使用容器名。
让我们将先前配置的Eureka设置为MicroService的注册表源,在微服务项目中配置:
spring.application.name=personsService
eureka.client.serviceUrl.defaultZone=http:<font><i>//eureka:9761/eureka</i></font><font>
eureka.client.healthcheck.enabled=<b>true</b>
eureka.instance.preferIpAddress=false
eureka.instance.hostname=service
ribbon.eureka.enabled=<b>true</b>
</font>
启动Docker中微服务:
(in the spring-microservice-service folder)
mvn clean install -DskipTests docker:build
docker images (verify your image was built)
docker run -it --link eureka:eureka --name service service/demo
docker ps (verify that service container is running)
使用“docker run”在service/demo容器将在别名eureka下看到Eureka服务器,让我们验证(运行以下命令):
docker exec -it service bash (<b>this</b> will connect you to the MicroService cont.)
ping eureka
输出:
ping eureka
PING eureka (172.17.0.2): 56 data bytes
64 bytes from 172.17.0.2: icmp_seq=0 ttl=64 time=0.269 ms
64 bytes from 172.17.0.2: icmp_seq=1 ttl=64 time=0.084 ms
网关容器
网关容器也需要链接到eureka和服务容器,因为eureka将以 http://service:8080 的形式提供ribbon负载平衡的MicroService地址。因此网关需要了解“服务”的含义。
application.properties:
server.port=8888
hystrix.command.invokePersonsMicroService.fallback.enabled=<b>true</b>
hystrix.command.invokePersonsMicroService.metrics.rollingStats.timeInMilliseconds=35000
hystrix.command.invokePersonsMicroService.circuitBreaker.sleepWindowInMilliseconds=5000
hystrix.command.invokePersonsMicroService.circuitBreaker.requestVolumeThreshold=6
hystrix.command.invokePersonsMicroService.circuitBreaker.errorThresholdPercentage=100
hystrix.command.invokePersonsMicroService.execution.isolation.strategy=THREAD
eureka.client.serviceUrl.defaultZone=http:<font><i>//eureka:9761/eureka</i></font><font>
eureka.client.healthcheck.enabled=<b>true</b>
eureka.client.registryFetchIntervalSeconds=5
eureka.client.fetch-registry=<b>true</b>
</font>
运行Docker中网关:
(in the spring-microservice-client folder)
mvn clean install -DskipTests docker:build
docker images (verify that image was built)
docker run -it --link eureka:eureka --link service:service --name client client/demo
运行镜像后,您应该再次看到重复输出:
Simple MicroService Feign invocation result :{<font>"persons"</font><font>:[{</font><font>"name"</font><font>:</font><font>"Tomas"</font><font>,</font><font>"surname"</font><font>:</font><font>"Kloucek"</font><font>,</font><font>"department"</font><font>:</font><font>"Programmer"</font><font>}]}
</font>
展示如何在所有节点之间创建基本的“桥梁”,这样他们就能看到对方!那么不需要手动Docker链接。
运行以下命令:
docker network create spring-cloud-network
既然我们现在有了一个网络,让我们把所有的节点放在其中
Maven编译和镜像构建与前面部分相同。
要启动eureka容器,请使用:
docker run -i -t --name eureka --net spring-cloud-network -p 9761:971 registry/demo
要启动MicroService容器,请使用:
docker run -it --name service --net spring-cloud-network service/demo
网格客户端,启动:
docker run -it --name client --net spring-cloud-network client/demo
让我们测试是否看到彼此:
docker exec -it service bash
60623256:/# ping eureka
PING eureka (172.19.0.2): 56 data bytes
64 bytes from 172.19.0.2: icmp_seq=0 ttl=64 time=0.085 ms
64 bytes from 172.19.0.2: icmp_seq=1 ttl=64 time=0.100 ms
64 bytes from 172.19.0.2: icmp_seq=2 ttl=64 time=0.110 ms
或
ping client
PING client (172.19.0.4): 56 data bytes
64 bytes from 172.19.0.4: icmp_seq=0 ttl=64 time=0.110 ms
64 bytes from 172.19.0.4: icmp_seq=1 ttl=64 time=0.106 ms
64 bytes from 172.19.0.4: icmp_seq=2 ttl=64 time=0.089 ms
如果您对每个容器彼此看到的拓扑结构都很好,那么基本的docker bridge网络就可以实现了,但请记住它只能在一个主机上运行。对于多个主机,您需要使用Overlay网络或服务网格
结合Docker运行Spring Cloud微服务的多种方式
标签:conf alt iso maven插件 sim .com 运行时 icm voc
原文地址:http://blog.51cto.com/14028890/2342121