cloudwatch.tf

resource "aws_cloudwatch_log_group" "backend" {
  name = "${module.global_variables.env}.ridi-pay-backend"

  tags = {
    Environment = module.global_variables.env
  }
}

data "aws_sns_topic" "cloudwatch_alarm" {
  name = "cloudwatch-alarm"
}

resource "aws_cloudwatch_metric_alarm" "pay_backend_alb_5xx_error" {
  count               = module.global_variables.is_prod ? 1 : 0
  alarm_name          = "pay-backend-alb-5xx-error"
  comparison_operator = "GreaterThanOrEqualToThreshold"
  evaluation_periods  = "1"
  metric_name         = "HTTPCode_ELB_5XX_Count"
  namespace           = "AWS/ApplicationELB"
  period              = "300"
  statistic           = "Sum"
  threshold           = "10"
  alarm_actions       = [data.aws_sns_topic.cloudwatch_alarm.arn]
  alarm_description   = "최근 5분 동안 ALB 5xx Error(Target 5xx 에러 제외) 10건 이상 발생"
  datapoints_to_alarm = 1
  dimensions = {
    LoadBalancer = aws_alb.ridi_pay_backend_fargate.arn_suffix
  }
}

resource "aws_cloudwatch_metric_alarm" "pay_backend_cpu_utilization" {
  count               = module.global_variables.is_prod ? 1 : 0
  alarm_name          = "pay-backend-cpu-utilization"
  comparison_operator = "GreaterThanOrEqualToThreshold"
  evaluation_periods  = "1"
  metric_name         = "CpuUtilization"
  namespace           = "AWS/ECS"
  period              = "300"
  statistic           = "Average"
  threshold           = "70"
  alarm_actions       = [data.aws_sns_topic.cloudwatch_alarm.arn]
  alarm_description   = "최근 5분 동안 pay-backend service 평균 CpuUtilization 70% 초과"
  datapoints_to_alarm = 1
  dimensions = {
    ClusterName = aws_ecs_cluster.ridi_pay_backend.name
    ServiceName = "ridi-pay-backend"
  }
}

resource "aws_cloudwatch_metric_alarm" "pay_backend_memory_utilization" {
  count               = module.global_variables.is_prod ? 1 : 0
  alarm_name          = "pay-backend-memory-utilization"
  comparison_operator = "GreaterThanOrEqualToThreshold"
  evaluation_periods  = "1"
  metric_name         = "MemoryUtilization"
  namespace           = "AWS/ECS"
  period              = "300"
  statistic           = "Average"
  threshold           = "85"
  alarm_actions       = [data.aws_sns_topic.cloudwatch_alarm.arn]
  alarm_description   = "최근 5분 동안 pay-backend service 평균 MemoryUtilization 85% 초과"
  datapoints_to_alarm = 1
  dimensions = {
    ClusterName = aws_ecs_cluster.ridi_pay_backend.name
    ServiceName = "ridi-pay-backend"
  }
}

resource "aws_cloudwatch_metric_alarm" "pay_backend_unhealthy_host_count" {
  count               = module.global_variables.is_prod ? 1 : 0
  alarm_name          = "pay-backend-unhealthy-host-count"
  comparison_operator = "GreaterThanOrEqualToThreshold"
  evaluation_periods  = "1"
  metric_name         = "UnHealthyHostCount"
  namespace           = "AWS/ApplicationELB"
  period              = "300"
  statistic           = "Maximum"
  threshold           = "1"
  alarm_actions       = [data.aws_sns_topic.cloudwatch_alarm.arn]
  alarm_description   = "ECS Task 1개 이상 Unhealthy"
  datapoints_to_alarm = 1
  dimensions = {
    LoadBalancer = aws_alb.ridi_pay_backend_fargate.arn_suffix
    TargetGroup  = aws_alb_target_group.ridi_pay_backend.arn_suffix
  }
}

resource "aws_cloudwatch_metric_alarm" "elasticache_001_cpu_utilization" {
  count               = module.global_variables.is_prod ? 1 : 0
  alarm_name          = "elasticache-001-cpu-utilization"
  comparison_operator = "GreaterThanOrEqualToThreshold"
  evaluation_periods  = "1"
  namespace           = "AWS/ElastiCache"
  metric_name         = "CPUUtilization"
  period              = "300"
  statistic           = "Average"
  threshold           = "45"
  alarm_actions       = [data.aws_sns_topic.cloudwatch_alarm.arn]
  alarm_description   = "최근 5분 동안 ElastiCache-001 평균 CPUUtilization vCPU당 45%, 총 90% 이상 / https://docs.aws.amazon.com/ko_kr/AmazonElastiCache/latest/red-ug/CacheMetrics.WhichShouldIMonitor.html 참고"
  datapoints_to_alarm = 1
  dimensions = {
    CacheClusterId = aws_elasticache_replication_group.redis[0].id
    CacheNodeId    = "0001"
  }
}

resource "aws_cloudwatch_metric_alarm" "elasticache_001_swap_usage" {
  count               = module.global_variables.is_prod ? 1 : 0
  alarm_name          = "elasticache-001-swap-usage"
  comparison_operator = "GreaterThanOrEqualToThreshold"
  evaluation_periods  = "1"
  namespace           = "AWS/ElastiCache"
  metric_name         = "SwapUsage"
  period              = "300"
  statistic           = "Maximum"
  threshold           = 50 * 1024 * 1024
  alarm_actions       = [data.aws_sns_topic.cloudwatch_alarm.arn]
  alarm_description   = "최근 5분 동안 ElastiCache-001 최대 SwapUsage 50MB 이상 / https://docs.aws.amazon.com/ko_kr/AmazonElastiCache/latest/red-ug/CacheMetrics.WhichShouldIMonitor.html 참고"
  datapoints_to_alarm = 1
  dimensions = {
    CacheClusterId = aws_elasticache_replication_group.redis[0].id
    CacheNodeId    = "0001"
  }
}

resource "aws_cloudwatch_metric_alarm" "elasticache_001_freeable_memory" {
  count               = module.global_variables.is_prod ? 1 : 0
  alarm_name          = "elasticache-001-freeable-memory"
  comparison_operator = "LessThanOrEqualToThreshold"
  evaluation_periods  = "1"
  namespace           = "AWS/ElastiCache"
  metric_name         = "FreeableMemory"
  period              = "300"
  statistic           = "Minimum"
  threshold           = 2 * 1024 * 1024 * 1024
  alarm_actions       = [data.aws_sns_topic.cloudwatch_alarm.arn]
  alarm_description   = "최근 5분 동안 ElastiCache-001 최소 FreeableMemory 2GB(30%) 이하"
  datapoints_to_alarm = 1
  dimensions = {
    CacheClusterId = aws_elasticache_replication_group.redis[0].id
    CacheNodeId    = "0001"
  }
}

resource "aws_cloudwatch_metric_alarm" "elasticache_001_is_master" {
  count               = module.global_variables.is_prod ? 1 : 0
  alarm_name          = "elasticache-001-is-master"
  comparison_operator = "LessThanThreshold"
  evaluation_periods  = "1"
  namespace           = "AWS/ElastiCache"
  metric_name         = "IsMaster"
  period              = "300"
  statistic           = "Minimum"
  threshold           = "1"
  alarm_actions       = [data.aws_sns_topic.cloudwatch_alarm.arn]
  alarm_description   = "최근 5분 동안 ElastiCache-001 Node Fail 발생"
  datapoints_to_alarm = 1
  dimensions = {
    CacheClusterId = aws_elasticache_replication_group.redis[0].id
    CacheNodeId    = "0001"
  }
}

resource "aws_cloudwatch_metric_alarm" "rds_master_error_log" {
  count               = module.global_variables.is_prod ? 1 : 0
  alarm_name          = "rds-master-error-log"
  comparison_operator = "GreaterThanOrEqualToThreshold"
  evaluation_periods  = "1"
  metric_name         = "IncomingLogEvents"
  namespace           = "AWS/Logs"
  period              = "300"
  statistic           = "Sum"
  threshold           = "1"
  alarm_actions       = [data.aws_sns_topic.cloudwatch_alarm.arn]
  alarm_description   = "최근 5분 동안 RDS error log 1건 이상 발생"
  datapoints_to_alarm = 1
  dimensions = {
    LogGroupName = "/aws/rds/instance/ridi-pay-prod-master/error"
  }
}

resource "aws_cloudwatch_metric_alarm" "rds_master_cpu_utilization" {
  count               = module.global_variables.is_prod ? 1 : 0
  alarm_name          = "rds-master-cpu-utilization"
  comparison_operator = "GreaterThanOrEqualToThreshold"
  evaluation_periods  = "1"
  metric_name         = "CPUUtilization"
  namespace           = "AWS/RDS"
  period              = "300"
  statistic           = "Average"
  threshold           = "70"
  alarm_actions       = [data.aws_sns_topic.cloudwatch_alarm.arn]
  alarm_description   = "최근 5분 동안 RDS master 평균 CPUUtilization 70% 이상"
  datapoints_to_alarm = 1
  dimensions = {
    DBInstanceIdentifier = "ridi-pay-prod-master"
  }
}

resource "aws_cloudwatch_metric_alarm" "rds_master_freeable_memory" {
  count               = module.global_variables.is_prod ? 1 : 0
  alarm_name          = "rds-master-freeable-memory"
  comparison_operator = "LessThanOrEqualToThreshold"
  evaluation_periods  = "1"
  namespace           = "AWS/RDS"
  metric_name         = "FreeableMemory"
  period              = "300"
  statistic           = "Minimum"
  threshold           = 100 * 1024 * 1024
  alarm_actions       = [data.aws_sns_topic.cloudwatch_alarm.arn]
  alarm_description   = "최근 5분 동안 RDS master 최소 FreeableMemory 100MB 이하"
  datapoints_to_alarm = 1
  dimensions = {
    DBInstanceIdentifier = "ridi-pay-prod-master"
  }
}

resource "aws_cloudwatch_metric_alarm" "rds_master_free_storage_space" {
  count               = module.global_variables.is_prod ? 1 : 0
  alarm_name          = "rds-master-free-storage-space"
  comparison_operator = "LessThanOrEqualToThreshold"
  evaluation_periods  = "1"
  namespace           = "AWS/RDS"
  metric_name         = "FreeStorageSpace"
  period              = "300"
  statistic           = "Minimum"
  threshold           = 75 * 1024 * 1024 * 1024
  alarm_actions       = [data.aws_sns_topic.cloudwatch_alarm.arn]
  alarm_description   = "최근 5분 동안 RDS master 최소 FreeStorageSpace 75GB(30%) 이하"
  datapoints_to_alarm = 1
  dimensions = {
    DBInstanceIdentifier = "ridi-pay-prod-master"
  }
}

resource "aws_cloudwatch_metric_alarm" "rds_master_swap_usage" {
  count               = module.global_variables.is_prod ? 1 : 0
  alarm_name          = "rds-master-swap-usage"
  comparison_operator = "GreaterThanOrEqualToThreshold"
  evaluation_periods  = "1"
  metric_name         = "SwapUsage"
  namespace           = "AWS/RDS"
  period              = "300"
  statistic           = "Maximum"
  threshold           = 100 * 1024 * 1024
  alarm_actions       = [data.aws_sns_topic.cloudwatch_alarm.arn]
  alarm_description   = "최근 5분 동안 최대 SwapUsage 100MB 이상"
  datapoints_to_alarm = 1
  dimensions = {
    DBInstanceIdentifier = "ridi-pay-prod-master"
  }
}

resource "aws_cloudwatch_metric_alarm" "rds_master_database_connections" {
  count               = module.global_variables.is_prod ? 1 : 0
  alarm_name          = "rds-master-database-connections"
  comparison_operator = "GreaterThanOrEqualToThreshold"
  evaluation_periods  = "1"
  metric_name         = "DatabaseConnections"
  namespace           = "AWS/RDS"
  period              = "300"
  statistic           = "Maximum"
  threshold           = "440"
  alarm_actions       = [data.aws_sns_topic.cloudwatch_alarm.arn]
  alarm_description   = "최근 5분 동안 최대 DatabaseConnections 440개(70%) 이상"
  datapoints_to_alarm = 1
  dimensions = {
    DBInstanceIdentifier = "ridi-pay-prod-master"
  }
}