EKS Alerting and Event Management Flow

EKS Alerting and Event Management Flow An architecture diagram generated by Archify. CloudWatch Metrics · Logs · metric- and log-based · Alert Sources CloudWatch Metrics · Logs metric- and log-based AWS Events · event-based · Alert Sources AWS Events event-based Prometheus Metrics · Loki Logs · incl. custom metrics · Alert Sources Prometheus Metrics · Loki Logs incl. custom metrics Kubernetes · Application Events · incl. OpenSearch logs · Alert Sources Kubernetes · Application Events incl. OpenSearch logs CloudWatch Alarms · alarm on threshold breach · Alert Processing CloudWatch Alarms alarm on threshold breach Amazon EventBridge · AWS event bus · Alert Processing Amazon EventBridge AWS event bus Prometheus Alertmanager · route · group · receiver · Alert Processing Prometheus Alertmanager route · group · receiver Event Router · custom routing · Alert Processing Event Router custom routing Amazon SNS · fan-out to Email · SQS subscriptions · Notification Channels Amazon SNS fan-out to Email · SQS subscriptions AWS Lambda · custom handler · Notification Channels AWS Lambda custom handler Slack · PagerDuty · Alertmanager receivers · Notification Channels Slack · PagerDuty Alertmanager receivers OpsGenie · on-call alerting · Notification Channels OpsGenie on-call alerting alarm-actions Alert Sources Alert Processing Notification Channels Legend Kubernetes / OSS AWS service Event · message bus External notification channel

CloudWatch Alarms

  • • Threshold alarms on ContainerInsights metrics such as pod_cpu_utilization
  • • --alarm-actions delivers notifications to an SNS topic

Prometheus Alertmanager

  • • route groups alerts via group_by · group_wait · repeat_interval
  • • receiver config routes to channels such as Slack