Terraform을 이용한 AWS Cloud Watch 모니터링 세팅

AWS Profile 지정

aws configure --profile <프로필 이름>

AWS Access Key ID [None]: <Access-Key>
AWS Secret Access Key [None]: <Secrey-Key>
Default region name [None]: ap-northeast-2
Default output format [None]: json

각 EC2에 CloudWatch Agent 설치

## CloudWatch Agent 설치
sudo yum install amazon-cloudwatch-agent

## CloudWatch Agent 설정 마법사 실행
sudo /opt/aws/amazon-cloudwatch-agent/bin/amazon-cloudwatch-agent-config-wizard

• 설정 파일 위치 /opt/aws/amazon-cloudwatch-agent/bin/config.json

설정 마법사 실행 로그

================================================================
= Welcome to the Amazon CloudWatch Agent Configuration Manager =
=                                                              =
= CloudWatch Agent allows you to collect metrics and logs from =
= your host and send them to CloudWatch. Additional CloudWatch =
= charges may apply.                                           =
================================================================
On which OS are you planning to use the agent?
1. linux
2. windows
3. darwin
default choice: [1]:
1
Trying to fetch the default region based on ec2 metadata...
I! imds retry client will retry 1 timesAre you using EC2 or On-Premises hosts?
1. EC2
2. On-Premises
default choice: [1]:
1
Which user are you planning to run the agent?
1. cwagent
2. root
3. others
default choice: [1]:
1
Do you want to turn on StatsD daemon?
1. yes
2. no
default choice: [1]:
2
Do you want to monitor metrics from CollectD? WARNING: CollectD must be installed or the Agent will fail to start
1. yes
2. no
default choice: [1]:
2
Do you want to monitor any host metrics? e.g. CPU, memory, etc.
1. yes
2. no
default choice: [1]:
1
Do you want to monitor cpu metrics per core?
1. yes
2. no
default choice: [1]:
1
Do you want to add ec2 dimensions (ImageId, InstanceId, InstanceType, AutoScalingGroupName) into all of your metrics if the info is available?
1. yes
2. no
default choice: [1]:
1
Do you want to aggregate ec2 dimensions (InstanceId)?
1. yes
2. no
default choice: [1]:
1
Would you like to collect your metrics at high resolution (sub-minute resolution)? This enables sub-minute resolution for all metrics, but you can customize for specific metrics in the output json file.
1. 1s
2. 10s
3. 30s
4. 60s
default choice: [4]:
4
Which default metrics config do you want?
1. Basic
2. Standard
3. Advanced
4. None
default choice: [1]:
1
Current config as follows:
{
    "agent": {
        "metrics_collection_interval": 60,
        "run_as_user": "cwagent"
    },
    "metrics": {
        "aggregation_dimensions": [
            [
                "InstanceId"
            ]
        ],
        "append_dimensions": {
            "AutoScalingGroupName": "${aws:AutoScalingGroupName}",
            "ImageId": "${aws:ImageId}",
            "InstanceId": "${aws:InstanceId}",
            "InstanceType": "${aws:InstanceType}"
        },
        "metrics_collected": {
            "disk": {
                "measurement": [
                    "used_percent"
                ],
                "metrics_collection_interval": 60,
                "resources": [
                    "*"
                ]
            },
            "mem": {
                "measurement": [
                    "mem_used_percent"
                ],
                "metrics_collection_interval": 60
            }
        }
    }
}
Are you satisfied with the above config? Note: it can be manually customized after the wizard completes to add additional items.
1. yes
2. no
default choice: [1]:
1
Do you have any existing CloudWatch Log Agent (http://docs.aws.amazon.com/AmazonCloudWatch/latest/logs/AgentReference.html) configuration file to import for migration?
1. yes
2. no
default choice: [2]:
2
Do you want to monitor any log files?
1. yes
2. no
default choice: [1]:
2
Do you want the CloudWatch agent to also retrieve X-ray traces?
1. yes
2. no
default choice: [1]:
2
Existing config JSON identified and copied to:  /opt/aws/amazon-cloudwatch-agent/etc/backup-configs
Saved config file to /opt/aws/amazon-cloudwatch-agent/bin/config.json successfully.
Current config as follows:
{
    "agent": {
        "metrics_collection_interval": 60,
        "run_as_user": "cwagent"
    },
    "metrics": {
        "aggregation_dimensions": [
            [
                "InstanceId"
            ]
        ],
        "append_dimensions": {
            "AutoScalingGroupName": "${aws:AutoScalingGroupName}",
            "ImageId": "${aws:ImageId}",
            "InstanceId": "${aws:InstanceId}",
            "InstanceType": "${aws:InstanceType}"
        },
        "metrics_collected": {
            "disk": {
                "measurement": [
                    "used_percent"
                ],
                "metrics_collection_interval": 60,
                "resources": [
                    "*"
                ]
            },
            "mem": {
                "measurement": [
                    "mem_used_percent"
                ],
                "metrics_collection_interval": 60
            }
        }
    }
}
Please check the above content of the config.
The config file is also located at /opt/aws/amazon-cloudwatch-agent/bin/config.json.
Edit it manually if needed.
Do you want to store the config in the SSM parameter store?
1. yes
2. no
default choice: [1]:
2
Program exits now.

terraform.tfvars 파일 생성 (각 프로젝트에 맞게)

cd ./terraform-cloudwatch

cat > terraform.tfvars << 'EOF'
# 프로젝트 설정
aws_region = "<리전>"
aws_profile = "<프로필 이름>"
project_name   = "<프로젝트 이름>"
dashboard_name = "<대시보드 이름>"

# EC2 인스턴스 정보
ec2_instances = [
  {
    instance_id = "<인스턴스 ID>"
    label       = "<LABEL>"
  },
  {
    instance_id = "<인스턴스 ID>"
    label       = "<LABEL>"
  }
]

# RDS Cluster 정보
rds_cluster = {
  cluster_id = "<클러스터 ID>"
  label       = "<LABEL>"
}


# RDS Writer 정보
rds_writer_instance = {
  instance_id = "<인스턴스 ID>"
  label       = "<LABEL>"
}

# RDS Reader 정보
rds_reader_instances = [
  {
    instance_id = "<인스턴스 ID>"
    label       = "<LABEL>"
  }
]


# CloudFront 배포 정보
cloudfront_distributions = [
  {
    distribution_id = "<distribution_id>"
    label           = "<LABEL>"
  },
  {
    distribution_id = "<distribution_id>"
    label           = "<LABEL>"
  }
]

# ElastiCache 설정
elasticache_cluster_id = "<elasticache_cluster_id>"

# ALB 설정
alb_name = "<app/~~>"
alb_target_groups = [
  {
    arn = "<arn>"
    label       = "<LABEL>"
  },
  {
    arn = "<arn>"
    label       = "<LABEL>"
  }
]


# 임계값 설정
unhealthy_host_threshold = 1
memory_usage_threshold = 80
cpu_usage_threshold = 50
disk_usage_threshold = 80
database_connections_threshold = 60
database_memory_threshold = 500000000
free_storage_threshold = 3000000000

# 태그
tags = {
  Project     = "<프로젝트명>"
  ManagedBy   = "terraform"
}
EOF

Terraform 실행

## 초기화
terraform init

## 계획 확인 (실제 반영 X)
terraform plan

## 적용 (현재 AWS Profile에 반영)
terraform apply

## 대시보드 URL 확인
terraform output dashboard_url

최종 대시보드

 

 

Github Repository

https://github.com/Darren4641/Terraform-AWS-CloudWatch/tree/main

GitHub - Darren4641/Terraform-AWS-CloudWatch: Terraform을 이용한 AWS Cloud Watch 모니터링 세팅