groups:
  - name: replication-worker.rules
    rules:
      - alert: ReplicationWorkerFrequentlyRestart
        expr: sum(changes(process_start_time_seconds{component="replication-worker"}[30m])) by (region, instance) >=4
        for: 5m
        labels:
          tier: serverless
          provider_type: aws-free-tier
          severity: critical
          response_class: page
          component: tikv
          oncall_team: tikv_unified
        annotations:
          message: |
            Replication worker frequently restart: region: {{ $labels.region }}, instance: {{ $labels.instance }}, values: {{ $value }}
          value: '{{ $value }}'
          region: '{{ $labels.region }}'
      - alert: ChangefeedCheckpointLagHigh
        expr: max(ticdc_owner_checkpoint_ts_lag) by (region, keyspace_id, changefeed) > 300
        for: 5m
        labels:
          tier: serverless
          provider_type: aws-free-tier
          severity: warning
          component: tikv
          oncall_team: tikv_unified
        annotations:
          message: |
            Changefeed checkpoint lag too high: region: {{ $labels.region }}, keyspace: {{ $labels.keyspace_id }}, changefeed: {{ $labels.changefeed }}, values: {{ $value }}
          value: '{{ $value }}'
          region: '{{ $labels.region }}'