Cloudwatch 客户端安装

采用AWS的amazon linux 2 标准AMI或基于这个AMI做定制的镜像安装的EC2实例默认已经集成了Cloudwatch客户端。
手工安装客户端可以参考官方文档：
https://docs.aws.amazon.com/AmazonCloudWatch/latest/monitoring/download-cloudwatch-agent-commandline.html

Cloudwatch配置样例

可以通过如下语句以向导方式生成一个配置模版：

  1. /opt/aws/amazon-cloudwatch-agent/bin/amazon-cloudwatch-agent-config-wizard

可以在这个模版基础上做修改，下面是一个定制后的配置样例：

  1. {
  2.     "agent": {
  3.         "metrics_collection_interval": 10,
  4.         "run_as_user": "root"
  5.     },
  6.     "metrics": {
  7.         "append_dimensions": {
  8.             "AutoScalingGroupName": "${aws:AutoScalingGroupName}",
  9.             "ImageId": "${aws:ImageId}",
  10.             "InstanceId": "${aws:InstanceId}",
  11.             "InstanceType": "${aws:InstanceType}"
  12.         },
  13.         "metrics_collected": {
  14.             "cpu": {
  15.                 "resources": [
  16.                     "*"
  17.                 ],
  18.                 "measurement": [
  19.                     {
  20.                         "name": "cpu_usage_idle",
  21.                         "unit": "Percent"
  22.                     },
  23.                     {
  24.                         "name": "cpu_usage_iowait",
  25.                         "unit": "Percent"
  26.                     },
  27.                     {
  28.                         "name": "cpu_usage_system",
  29.                         "unit": "Percent"
  30.                     },
  31.                     {
  32.                         "name": "cpu_usage_user",
  33.                         "unit": "Percent"
  34.                     },
  35.                     {
  36.                         "name": "cpu_usage_nice",
  37.                         "unit": "Percent"
  38.                     },
  39.                     "cpu_usage_guest"
  40.                 ],
  41.                 "totalcpu": false,
  42.                 "metrics_collection_interval": 10
  43.             },
  44.             "netstat": {
  45.                 "measurement": [
  46.                     "tcp_established",
  47.                     "tcp_syn_sent",
  48.                     "tcp_close"
  49.                 ],
  50.                 "metrics_collection_interval": 60
  51.             },
  52.             "disk": {
  53.                 "measurement": [
  54.                     "used_percent",
  55.                     "disk_inodes_free"
  56.                 ],
  57.                 "resources": [
  58.                     "/",
  59.                     "/home/t4",
  60.                     "/emr",
  61.                     "/mnt",
  62.                     "/mnt1",
  63.                     "/var/log"
  64.                 ],
  65.                 "metrics_collection_interval": 60,
  66.                 "drop_device": true
  67.             },
  68.             "processes": {
  69.                 "measurement": [
  70.                     "running",
  71.                     "sleeping",
  72.                     "dead"
  73.                 ]
  74.             },
  75.             "mem": {
  76.                 "measurement": [
  77.                     "mem_used_percent",
  78.                     "mem_used",
  79.                     "mem_cached",
  80.                     "mem_total"
  81.                 ],
  82.                 "metrics_collection_interval": 10
  83.             }
  84.         }
  85.     }
  86. }

配置推送到 ssm的parameter store

可以将上述配置保存成文件，然后推送到ssm的parameter store，方便各个机器共享配置文件，实现统一配置管理。

  1. aws ssm put-parameter --name "parameter name" --type "String" --value file://configuration_file_pathname

配置生效

在EC2上执行如下语句将parameter store上的配置文件下载到本地，并重启服务生效：

  1. /opt/aws/amazon-cloudwatch-agent/bin/amazon-cloudwatch-agent-ctl -a fetch-config -m ec2 -s -c ssm:AmazonCloudWatch-linux