apiVersion: monitoring.coreos.com/v1
kind: PrometheusRule
metadata:
  name: dawarich-valkey
  labels:
    helm.sh/chart: valkey-0.9.3
    app.kubernetes.io/name: valkey
    app.kubernetes.io/instance: dawarich
    app.kubernetes.io/version: "9.0.3"
    app.kubernetes.io/managed-by: Helm
    app.kubernetes.io/part-of: valkey
spec:
  groups:
    - name: dawarich-valkey
      rules:
        - alert: ValkeyDown
          annotations:
            description: Valkey instance {{ $labels.instance }} is down.
            summary: Valkey instance {{ $labels.instance }} down
          expr: |
            redis_up{service="dawarich-valkey-metrics"} == 0
          for: 2m
          labels:
            severity: error
        - alert: ValkeyMemoryHigh
          annotations:
            description: |
              Valkey instance {{ $labels.instance }} is using {{ $value }}% of its available memory.
            summary: Valkey instance {{ $labels.instance }} is using too much memory
          expr: |
            redis_memory_used_bytes{service="dawarich-valkey-metrics"} * 100
            /
            redis_memory_max_bytes{service="dawarich-valkey-metrics"}
            > 90 <= 100
          for: 2m
          labels:
            severity: error
        - alert: ValkeyKeyEviction
          annotations:
            description: |
              Valkey instance {{ $labels.instance }} has evicted {{ $value }} keys in the last 5 minutes.
            summary: Valkey instance {{ $labels.instance }} has evicted keys
          expr: |
            increase(redis_evicted_keys_total{service="dawarich-valkey-metrics"}[5m]) > 0
          for: 1s
          labels:
            severity: error