所有组件都以容器形式启动，部分启动文件参考prometheus for swarm

1. 部署Prometheus

   1. 编写启动文件

      $ mkdir -p /opt/k8s/prometheus/conf
      $ cd /opt/k8s/prometheus/
      $ cat > prome-stack.yml<<EOF
      version: "3"
      services:
        prometheus: 
          image: prom/prometheus:v2.16.0
          ports:
            - "9090:9090"
          volumes:
            - ./conf/:/etc/prometheus/
            - prometheus_data:/prometheus
          command:
            - '--config.file=/etc/prometheus/prometheus.yml'
            - '--storage.tsdb.path=/prometheus'
            - '--web.console.libraries=/usr/share/prometheus/console_libraries'
            - '--web.console.templates=/usr/share/prometheus/consoles'
          networks: 
            - mcsas-network
          deploy:
            replicas: 1 
            restart_policy:
              condition: on-failure
            placement:
              constraints:
                - node.role == manager
      networks:
        mcsas-network: 
          external: true
      volumes:
        prometheus_data: {}
      EOF
      

   2. 编辑配置文件

      $ cd /opt/k8s/prometheus/prom/conf/
      $ cat > prometheus.yml<<EOF
      global:
        scrape_interval: 15s 
        evaluation_interval: 15s
      scrape_configs:
        - job_name: 'prometheus'
          static_configs:
          - targets: ['localhost:9090']
        - job_name: 'springboot'
          metrics_path: /actuator/prometheus
          file_sd_configs:
          - files:
            - /etc/prometheus/service.yaml
        - job_name: 'node-exporter'
          scrape_interval: 5s
          dns_sd_configs:
          - names:
            - 'tasks.node-exporter'
            type: 'A'
            port: 9100    
        - job_name: 'cadvisor'
          scrape_interval: 5s
      
          dns_sd_configs:
          - names:
            - 'tasks.cadvisor'
            type: 'A'
            port: 8080        
      EOF
      

      • 对node-exporter、cadvisor采用dns服务发现形式
      • 对于系统应用采用file_sd_configs，通过conf/service.yaml中配置，来是prometheus对我们提供的服务进行监控
      • 因为Prometheus没有专门针对swarm的服务发现组件，需要手动向file_sd_configs对应的文件中追加，Prometheus官方上有一个方案，具体可参考prometheus-swarm-discovery

   3. 启动prometheus

      $ cd /opt/k8s/prometheus/prom
      $ docker stack deploy -c prome-stack.yml prom
      

2. 部署node-exporter

   Node-Exporter并不是为了Mac平台设计的，在Mac上运行时不会正确收集系统相关的信息，如果平台是Mac，不要部署这个组件

   $ cd /opt/k8s/prometheus
   $ cat > node-exporter-stack.yml<<EOF
   version: "3"
   services:
     node-exporter:
       image: quay.azk8s.cn/prometheus/node-exporter:v0.18.1
       volumes:
         - /proc:/host/proc:ro
         - /sys:/host/sys:ro
         - /:/rootfs:ro
       command:
         - '--path.procfs=/host/proc' 
         - '--path.sysfs=/host/sys'
         - --collector.filesystem.ignored-mount-points
         - "^/(sys|proc|dev|host|etc|rootfs/var/lib/docker/containers|rootfs/var/lib/docker/overlay2|rootfs/run/docker/netns|rootfs/var/lib/docker/aufs)($$|/)"
       #ports:
       #  - 9100:9100
       networks:
         - mcsas-network
       deploy:
         mode: global
         restart_policy:
           condition: on-failure
   networks:
     mcsas-network: 
       external: true
   EOF
   

   启动node-exporter

   $ cd /opt/k8s/prometheus
   $ docker stack deploy -c node-exporter-stack.yml node
   

3. 部署cadvisor

   $ cd /opt/k8s/prometheus
   $ cat > cadvisor-stack.yml<<EOF
   version: "3"
   services:
     cadvisor:
       image: gcr.azk8s.cn/google_containers/cadvisor:v0.35.0 
       volumes:
         - /:/rootfs:ro
         - /var/run:/var/run:rw
         - /sys:/sys:ro
         - /var/lib/docker/:/var/lib/docker:ro
       #ports:
       #  - 8080:8080
       networks:
         - mcsas-network
       deploy:
         mode: global
         restart_policy:
           condition: on-failure
   
   networks:
     mcsas-network: 
       external: true
   EOF
   

   • 关于镜像：google/cadvisor已经不推荐再用，新镜像已不再更新，使用gcr.io/google-containers/cadvisor，但是国内没发下载，更换成从gcr.azk8s.cn下载

   启动cadvisor

   $ docker stack deploy -c cadvisor-stack.yml cadvisor
   

4. 部署grafana

   $ cd /opt/k8s/prometheus
   $ cat > grafana-stack.yml<<EOF
   version: "3"
   services:
   
     grafana:
       image: grafana/grafana:6.6.2
       volumes:
         - grafana-data:/var/lib/grafana
       deploy:
         replicas: 1
         restart_policy:
           condition: on-failure
         resources:
           limits:
             cpus: "0.2"
             memory: 200M
       ports:
         - 3000:3000
       networks:
         - mcsas-network
   
   volumes:
     grafana-data: {}
   
   networks:
     mcsas-network:
       external: true
   EOF
   

   启动grafana

   $ docker stack deploy -c grafana-stack.yml grafana
   

部署组件全部完成，关于在grafana中配置dashboard进行指标监控的具体步骤，参考Prometheus grafana安装