repository·main·Indexed 27 days ago
https://github.com/prometheus-community/helm-chartsOfficial Kubernetes Helm charts maintained by the Prometheus community, providing standardized deployment templates for Prometheus-related components. Includes charts for kube-prometheus-stack, Alertmanager, jiralert, and alertmanager-snmp-notifier, with support for OCI registries and traditional Helm repositories.
The chart is available through two distribution methods:
oci://ghcr.io/prometheus-community/charts/prometheus-conntrack-stats-exporterhttps://prometheus-community.github.io/helm-charts (chart name: prometheus-conntrack-stats-exporter)The chart is available through two distribution methods:
oci://ghcr.io/prometheus-community/charts/prometheus-windows-exporterhttps://prometheus-community.github.io/helm-charts (chart name: prometheus-windows-exporter)The chart is available through two distribution methods:
oci://ghcr.io/prometheus-community/charts/prometheus-sql-exporterhttps://prometheus-community.github.io/helm-charts (chart name: prometheus-sql-exporter)Use the helm upgrade command to update your existing release. Use the --install flag to ensure the chart is installed if it doesn't already exist.
helm upgrade [RELEASE_NAME] [CHART] --installThe chart was renamed from stable/elasticsearch-exporter to prometheus-elasticsearch-exporter. To maintain existing resource names (like Service or Deployment) during an upgrade, use fullnameOverride and nameOverride.
Example for a release named my-exporter:
helm upgrade my-exporter . --set fullnameOverride=my-exporter-elasticsearch-exporter --set nameOverride=elasticsearch-exporterThe chart is distributed as an OCI Artifact. You can install it using the following command. Ensure you provide a [RELEASE_NAME] and configure the necessary MongoDB connection settings (see Configuration section).
helm install [RELEASE_NAME] oci://ghcr.io/prometheus-community/charts/prometheus-mongodb-exporterTo migrate and change the name prefix (which requires downtime), follow these steps. This guide assumes the deployment is in the monitoring namespace.
Patch the existing PV to use a Retain reclaim policy:
kubectl patch pv/<PersistentVolume name> -p '{"spec":{"persistentVolumeReclaimPolicy":"Retain"}}'Uninstall the old release and delete the PVC:
helm uninstall prometheus-operator -n monitoring
kubectl delete pvc/<PersistenceVolumeClaim name> -n monitoringNote: You must also manually delete the prometheus-operator-kubelet service in the kube-system namespace.
Make the PV Available: Remove the spec.claimRef from the PV so it can be re-bound:
kubectl patch pv/<PersistentVolume name> --type json -p='[{"op": "remove", "path": "/spec/claimRef"}]' -n monitoringFresh Installation: Install kube-prometheus-stack ensuring the volumeClaimTemplate (storage size, access modes, and storageClassName) matches your old configuration.
Important: Ensure the new pods are scheduled in the same Availability Zone (AZ) as the old PV. You can enforce this using prometheus.prometheusSpec.nodeSelector in your values.yaml:
nodeSelector:
failure-domain.beta.kubernetes.io/zone: east-west-1aUpgrade an existing release or install it if it does not exist.
Command:
helm upgrade [RELEASE_NAME] [CHART] --installThis upgrade updates Prometheus-Operator to v0.62.0, Prometheus to v2.41.0, and Thanos to v0.30.1. You must update the CRDs using kubectl apply --server-side before applying the Helm upgrade.
prometheusOperator.admissionWebhooks.failurePolicy, this value is now always used even when .prometheusOperator.admissionWebhooks.patch.enabled is true (the default).prometheusOperator.image.tag and prometheusOperator.prometheusConfigReloader.image.tag are now empty by default; the Chart.yaml appVersion field is used instead.kubectl apply --server-side -f https://raw.githubusercontent.com/prometheus-operator/prometheus-operator/v0.62.0/example/prometheus-operator-crd/monitoring.coreos.com_alertmanagerconfigs.yaml
kubectl apply --server-side -f https://raw.githubusercontent.com/prometheus-operator/prometheus-operator/v0.62.0/example/prometheus-operator-crd/monitoring.coreos.com_alertmanagers.yaml
kubectl apply --server-side -f https://raw.githubusercontent.com/prometheus-operator/prometheus-operator/v0.62.0/example/prometheus-operator-crd/monitoring.coreos.com_podmonitors.yaml
kubectl apply --server-side -f https://raw.githubusercontent.com/prometheus-operator/prometheus-operator/v0.62.0/example/prometheus-operator-crd/monitoring.coreos.com_probes.yaml
kubectl apply --server-side -f https://raw.githubusercontent.com/prometheus-operator/prometheus-operator/v0.62.0/example/prometheus-operator-crd/monitoring.coreos.com_prometheuses.yaml
kubectl apply --server-side -f https://raw.githubusercontent.com/prometheus-operator/prometheus-operator/v0.62.0/example/prometheus-operator-crd/monitoring.coreos.com_prometheusrules.yaml
kubectl apply --server-side -f https://raw.githubusercontent.com/prometheus-operator/prometheus-operator/v0.62.0/example/prometheus-operator-crd/monitoring.coreos.com_servicemonitors.yaml
kubectl apply --server-side -f https://raw.githubusercontent.com/prometheus-operator/prometheus-operator/v0.62.0/example/prometheus-operator-crd/monitoring.coreos.com_thanosrulers.yamlkubeStateMetrics.serviceMonitor.namespaceOverride has been removed. Use kube-state-metrics.namespaceOverride instead.You can locally test the kube-prometheus-stack setup using Minikube. The configuration in the hack/minikube directory provides a way to set up components and create a working etcd scrape configuration.
To get a local working Minikube cluster, run the commands contained in cmd.sh in the exact sequence listed within the script.
Windows Users: If you are running on Windows, locate the commented-out section within the script and add those specific flags/commands to your minikube command to ensure compatibility.