任务：

1、docker服务加监控检查
2、检查的结果告警到飞书/钉钉等有机器人的群

docker服务加监控检查

原理就是每10s检查一次访问https://api.jd.com/routerjson?的网络码是否200，如果连续3次超时5s都没反应或者网络码不是200，则健康检查不通过，报unhealthy，反之healthy。

修改yaml，添加healthcheck部分。

version: '3.4'

services:
  <service name>:
    container_name: <container_name>
    restart: on-failure
    image: <iamge>
    networks:
      - backend_net
    environment:
      TZ: "Asia/Shanghai"
      JVM_XMX: "1024m"
      JVM_XMS: "1024M"
      JVM_XMN: "512M"
      JAR_NAME: "xxx.jar"
      CUSTOM_SEARCH_LOCATION: "/opt/bootstrap.properties"
    volumes:
      - $PWD/xxx/bootstrap-v2.properties:/opt/bootstrap.properties:ro
      - $PWD/xxx/logs:/opt/logs:rw
      - /etc/localtime:/etc/localtime:ro
    ports:
      - 8082:8080
    healthcheck:
      test: ["CMD-SHELL", "curl -s -o /dev/null -w '%{http_code}' https://api.jd.com/routerjson? | grep -q 200"]
      interval: 10s
      timeout: 5s
      retries: 3

networks:
  backend_net:
    ipam:
      driver: default
      config:
        - subnet: 172.16.248.0/24

[root@a ]# cat /root/alarm.sh
#!/bin/bash

api=https://open.feishu.cn/open-apis/bot/v2/hook/xxx  #飞书机器人webhook 地址
time=$(date "+%H:%M:%S")
restart_time=$(docker ps  |grep <service name>|awk '{print $10 $11}')
status=$(docker ps  |grep <service name>|grep -o '(.*)')
#status="unhealthy"
content="app:<service name>"  #要发送的信息

content="报警时间:$time\n报警事件:容器30s内3次访问https://api.jd.com/routerjson失败!\n报警规则:每10s访问一次url，连续3次没返回网络码200或超时5秒\n\n$content\n重启时间:$restart_time前\n状态:$status"
#换行用\n 空格用\t


function post(){
curl -X POST \
  $api \
  -H 'Content-Type: application/json' \
  -d '{
    "msg_type": "post",
    "content": {
        "post": {
            "zh_cn": {
                "title": "连接异常告警！",
                "content": [
                    [
                        {
                            "tag": "text",
                            "un_escape": true,
                            "text": "'$content'"
                        }
                    ],
                    [

                    ]
                ]
            }
        }
    }
}'
}
if [[ $status =~ "unhealthy" ]];then post;fi;

[root@a]# crontab -l
#告警服务不健康
*/5 * * * *  sh  /root/alarm.sh

告警效果图如下