Loki Helm 5.14.1 ruler not evaluating

Grafana Loki
1

Hello, can someone help me figure it out where is the problem in my loki ruler config? I am using s3 as chunks storage but want to use local for rules.

  • Using single tenant.
  • Using helm chart loki-5.14.1.

Below my ruler block config:

ruler:
  alertmanager_url: http://alertmanager.monitoring.svc.cluster.local:9093
  enable_alertmanager_v2: true
  enable_api: true
  ring:
    kvstore:
      store: inmemory
  rule_path: "/rules"
  storage:
    local:
      directory: /var/loki/rules
    type: local

On the read component in values.yaml i have the volumes mounted:

  extraVolumeMounts: 
    - name: loki-rules
      mountPath: "/var/loki/rules/fake"

  extraVolumes:
    - name: loki-rules
      configMap:
        name: loki-alerting-rules

Have a configmap deployed and mapped to the rules.yaml:

cat /loki/rules/fake/rules.yaml
groups:
  - name: high_error_rate
    rules:
      - alert: LokiHighErrorRate
        expr: sum(rate({cluster="loki"} | logfmt | level="info"[1m])) by (container) > 0
        for: 1m
        labels:
            severity: warning
        annotations:
            summary: Loki error per minute rate > 0
            message: 'Loki warning per minute rate > 0 container:"{{ $labels.container }}"'

There are no errors on logs, but rule seems not to be evaluated and nothing is writeen in rule_path “/rules” and no alarms are raised.

Can someone help on this?

Thanks

2

Managed to fix it, the rules.yaml need to be mapped to loki-backend instead of loki-read. It should have been changed recently and documentation not updated.

3

Can you please let me know where this parameter needs to be updated?