Installation¶

Prerequisites¶

Kubernetes 1.19+
kubectl configured for your cluster
Helm 3.x (for Helm installation)

Helm (Recommended)¶

1. Apply CRDs¶

CRDs are managed separately from the Helm chart to avoid accidental deletion on helm uninstall:

kubectl apply -f crds/

2. Install the Operator¶

helm install backup-operator ./charts/backup-operator \
  -n bnerd-backup-system --create-namespace

3. Verify¶

kubectl get pods -n bnerd-backup-system -l app.kubernetes.io/name=backup-operator

Expected output:

NAME                                    READY   STATUS    RESTARTS   AGE
backup-operator-xxxxxxxxxx-xxxxx        1/1     Running   0          30s

Key Helm Values¶

Parameter	Default	Description
`image.repository`	`registry.bnerd.com/public/bnerd-backup-operator`	Operator image
`image.tag`	`latest`	Image tag
`image.pullPolicy`	`Always`	Pull policy
`replicaCount`	`1`	Number of replicas
`rbac.create`	`true`	Create RBAC resources
`rbac.clusterWide`	`true`	Use ClusterRole (cluster-wide access)
`resources.requests.cpu`	`100m`	CPU request
`resources.requests.memory`	`128Mi`	Memory request
`resources.limits.cpu`	`500m`	CPU limit
`resources.limits.memory`	`512Mi`	Memory limit
`metrics.enabled`	`true`	Enable Prometheus metrics on port 8080
`metrics.serviceMonitor.enabled`	`false`	Create ServiceMonitor for Prometheus Operator
`metrics.prometheusRule.enabled`	`false`	Create PrometheusRule for alerting

RBAC Modes¶

Cluster-wide (default) -- the operator can manage backups across all namespaces:

helm install backup-operator ./charts/backup-operator \
  --set rbac.clusterWide=true

Namespace-scoped -- the operator only manages backups in its own namespace:

helm install backup-operator ./charts/backup-operator \
  --set rbac.clusterWide=false

Enable Monitoring¶

To create a ServiceMonitor for automatic Prometheus scraping:

helm install backup-operator ./charts/backup-operator \
  --set metrics.serviceMonitor.enabled=true

To also enable pre-built alerting rules:

helm install backup-operator ./charts/backup-operator \
  --set metrics.serviceMonitor.enabled=true \
  --set metrics.prometheusRule.enabled=true

See the Monitoring and Alerting guide for details.

Upgrade¶

helm upgrade backup-operator ./charts/backup-operator \
  -n bnerd-backup-system

After upgrading the operator, all VolumeBackup CronJobs are automatically reconciled within 5 minutes to pick up any changes.

Uninstall¶

helm uninstall backup-operator -n bnerd-backup-system

Note

CRDs and existing backup resources are not deleted when uninstalling the Helm chart. Remove CRDs manually with kubectl delete -f crds/ if desired.

Kustomize¶

kubectl apply -k manifests/

This deploys the operator into the bnerd-backup-system namespace with default settings.

To customise the image tag, edit manifests/kustomization.yaml:

images:
  - name: registry.bnerd.com/public/bnerd-backup-operator
    newTag: v0.9.1

Endpoints¶

Once running, the operator exposes two HTTP endpoints:

Port	Path	Purpose
8080	`/metrics`	Prometheus metrics
8081	`/healthz`	Liveness probe
8081	`/readyz`	Readiness probe (healthy if reconciled in the last 10 minutes)