Files
deprecated-helm-charts/stable/prometheus-operator/hack
Vasily Sliouniaev fbe9c7e097 [stable/prometheus-operator] Remove CRD hooks from CI entirely (#18271)
* Remove CRD hooks from CI entirely

Installing the operator by itself will create CRDs in the cluster
idempotently. We can then run the rest of the tests safely

Signed-off-by: vas <vasily.sliouniaev@jet.com>

* Remove irrelevant value

This cannot be relied upon because "CRD deletion is slow". Despite this
being a closed bug it's still an issue https://github.com/helm/helm/issues/4489

Signed-off-by: vas <vasily.sliouniaev@jet.com>
2019-10-28 05:31:26 -07:00
..

prometheus-operator hacks

sync_prometheus_rules.py

This script generates prometheus rules set for alertmanager from any properly formatted kubernetes yaml based on defined input, splitting rules to separate files based on group name.

Currently following imported:

  • coreos/kube-prometheus rules set
    • In order to modify these rules:

      jb update
      make generate-in-docker
      
      • prepare and merge PR with imported changes into coreos/kube-prometheus master and/or release branch
      • run sync_prometheus_rules.py inside your fork of this repo
      • send PR with changes to this repo
  • etcd-io/etc rules set
    • In order to modify these rules:
      • prepare and merge PR into etcd-io/etcd repo
      • run sync_prometheus_rules.py inside your fork of this repo
      • send PR with changes to this repo

sync_grafana_dashboards.py

This script generates grafana dashboards from json files, splitting them to separate files based on group name.

Currently following imported:

  • coreos/kube-prometheus dashboards
    • In order to modify these dashboards:

      jb update
      make generate-in-docker
      
      • prepare and merge PR with imported changes into coreos/kube-prometheus master and/or release branch
      • run sync_grafana_dashboards.py inside your fork of this repo
      • send PR with changes to this repo
  • etcd-io/etc dashboard
    • In order to modify this dashboard:
      • prepare and merge PR into etcd-io/etcd repo
      • run sync_grafana_dashboards.py inside your fork of this repo
      • send PR with changes to this repo

CoreDNS dashboard is the only dashboard which is maintained in this repo and can be changed without import.