Minio Prometheus Alert Rules

groups:
- name: EmbeddedExporter
  rules:
    - alert: MinioClusterErasureSetQuorumLost
      expr: minio_cluster_health_erasure_set_status < 1
      for: 0m
      labels:
        severity: critical
      annotations:
        summary: Minio cluster erasure set quorum lost (instance {{ $labels.instance }})
        description: "Minio erasure set has lost quorum, meaning MinIO can no longer guarantee reads or writes for objects in this pool/set until enough drives are restored.\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"
    
    - alert: MinioClusterDiskOffline
      expr: minio_cluster_drive_offline_total > 0
      for: 0m
      labels:
        severity: critical
      annotations:
        summary: Minio cluster disk offline (instance {{ $labels.instance }})
        description: "Minio cluster disk is offline\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"
    
    - alert: MinioNodeDiskOffline
      expr: minio_cluster_nodes_offline_total > 0
      for: 0m
      labels:
        severity: critical
      annotations:
        summary: Minio node disk offline (instance {{ $labels.instance }})
        description: "Minio cluster node is offline\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"
    
      # Uses usable capacity (post erasure-coding) rather than raw capacity, so the percentage
      # reflects how much data can actually still be stored, not the physical disk bytes
      # reserved for parity.
    - alert: MinioDiskSpaceUsage
      expr: minio_cluster_capacity_usable_free_bytes / minio_cluster_capacity_usable_total_bytes * 100 < 10 and minio_cluster_capacity_usable_total_bytes > 0
      for: 0m
      labels:
        severity: warning
      annotations:
        summary: Minio disk space usage (instance {{ $labels.instance }})
        description: "Minio available free space is low (< 10%)\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"
    
    - alert: MinioKMSUnavailable
      expr: minio_cluster_kms_online == 0
      for: 0m
      labels:
        severity: critical
      annotations:
        summary: Minio KMS unavailable (instance {{ $labels.instance }})
        description: "Minio KMS backend is offline; encryption and decryption of SSE-KMS objects will fail until KMS connectivity is restored.\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"

wget https://raw.githubusercontent.com/samber/awesome-prometheus-alerts/refs/heads/master/dist/rules/minio/embedded-exporter.yml

- alert: MinioClusterErasureSetQuorumLost
  expr: minio_cluster_health_erasure_set_status < 1
  for: 0m
  labels:
    severity: critical
  annotations:
    summary: Minio cluster erasure set quorum lost (instance {{ $labels.instance }})
    description: "Minio erasure set has lost quorum, meaning MinIO can no longer guarantee reads or writes for objects in this pool/set until enough drives are restored.\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"

- alert: MinioClusterDiskOffline
  expr: minio_cluster_drive_offline_total > 0
  for: 0m
  labels:
    severity: critical
  annotations:
    summary: Minio cluster disk offline (instance {{ $labels.instance }})
    description: "Minio cluster disk is offline\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"

- alert: MinioNodeDiskOffline
  expr: minio_cluster_nodes_offline_total > 0
  for: 0m
  labels:
    severity: critical
  annotations:
    summary: Minio node disk offline (instance {{ $labels.instance }})
    description: "Minio cluster node is offline\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"

  # Uses usable capacity (post erasure-coding) rather than raw capacity, so the percentage
  # reflects how much data can actually still be stored, not the physical disk bytes
  # reserved for parity.
- alert: MinioDiskSpaceUsage
  expr: minio_cluster_capacity_usable_free_bytes / minio_cluster_capacity_usable_total_bytes * 100 < 10 and minio_cluster_capacity_usable_total_bytes > 0
  for: 0m
  labels:
    severity: warning
  annotations:
    summary: Minio disk space usage (instance {{ $labels.instance }})
    description: "Minio available free space is low (< 10%)\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"

- alert: MinioKMSUnavailable
  expr: minio_cluster_kms_online == 0
  for: 0m
  labels:
    severity: critical
  annotations:
    summary: Minio KMS unavailable (instance {{ $labels.instance }})
    description: "Minio KMS backend is offline; encryption and decryption of SSE-KMS objects will fail until KMS connectivity is restored.\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"