Charts: How to add additional prometheus rules?

[boxcee]

Is this a request for help?:
Yes

Version of Helm and Kubernetes:
kubernetes client version: v1.12.3
kubernetes server version: v1.12.3
helm client version: v2.12.2+g7d2b0c7

Which chart:
stable/prometheus-operator:4.3.3

What happened:
Error while upgrading with newer values:

Command returned with exit code: 
. And error message: Error: UPGRADE FAILED:
 failed to create resource: PrometheusRule.monitoring.coreos.com "prometheus-operator-"
 is invalid: metadata.name: Invalid value: "prometheus-operator-": a DNS-1123 subdomain
 must consist of lower case alphanumeric characters, '-' or '.', and must start and end with
 an alphanumeric character
 (e.g. 'example.com', regex used for validation is '[a-z0-9]([-a-z0-9]*[a-z0-9])?(\.[a-z0-9]([-a-z0-9]*[a-z0-9])?)*')

ERROR: Job failed: command terminated with exit code 1

What you expected to happen:
I expect the additionalPrometheusRules to be created correctly without an error.

How to reproduce it (as minimally and precisely as possible):
Use this in your values.yaml:

additionalPrometheusRules:
  - groups:
    - name: elastic.rules
      rules:
      - alert: ElasticsearchDiskWarning
        annotations:
        description: Elastic Search disk full Warning
        summary: Elastic Search disk full Warning
        expr: 100 - es_fs_total_free_bytes{pod=~".*data.*"} / es_fs_total_total_bytes{pod=~".*data.*"}  * 100 > 80
        for: 1m
        labels:
          severity: warning
    - name: onesearch.rules
      rules:
      - alert: OneSearchClusterWarn
        annotations:
        description: The cluster {{ $labels.namespace }}.{{ $labels.cluster }} is yellow. Please check!
        summary: OneSearch cluster warning
        expr: es_cluster_status == 1
        for: 1m
        labels:
          severity: warn

[vsliouniaev]

Your format for the value of additionalPrometheusRules doesn't look right to me:
https://github.com/helm/charts/blob/master/stable/prometheus-operator/values.yaml#L47

[boxcee]

Yes, you are right. I was looking at the documentation here: https://github.com/coreos/prometheus-operator/blob/master/Documentation/api.md#prometheusrulespec. It only says groups there. Thanks!

[vsliouniaev]

The issue at the chart level is that every one of these prometheusrule objects needs to get its own name when it is provisioned, that's why the extra - name: is needed

[ori-stoliar]

Hi Guys,

I am adding the following configuration that I am passing to prometheus-operator chart:

additionalPrometheusRules:
    - name: example
        groups:
        - name: mygroup
            rules:
            - record: ContainerRestarted
            expr: count(sum by (pod)(delta(kube_pod_container_status_restarts_total[15m]) > 0))
            labels:
                severity: warning
            annotations:
                summary: "Container {{ $labels.container }} in Pod {{$labels.namespace}}/{{$labels.pod}} restarting"

When accessing Prometheus UI and searching for the alert it doesn't exist, also I am not able to find this alert in the configuration files physically when going into the Prometheus container.
The Prometheus configuration is being loaded successfully - there are no errors.
Any ideas?