New to KubeDB? Please start here.

Elasticsearch Branch Configuration

These optional examples reuse elasticsearch-source from the same-cluster tutorial. Keep that source and its namespaces before cleanup. Run one optional example at a time, deleting its Branch before creating the next to limit resource usage.

Storage and resource overrides

target.storageClassName selects the clone’s StorageClass; it must resolve the source snapshot on the same backend. target.resources sets target container CPU and memory requests and limits. It does not resize storage. The examples set both explicitly.

Select a snapshot class

Create a compatible ceph-rbd-snapclass before applying this example:

Save the following as elasticsearch-snapshot.yaml:

apiVersion: courier.kubedb.com/v1alpha1
kind: Branch
metadata:
  name: elasticsearch-snapshot
  namespace: demo
spec:
  source:
    databaseRef:
      apiGroup: kubedb.com
      kind: Elasticsearch
      name: elasticsearch-source
    namespace: demo
  target:
    namespace: dev
    name: elasticsearch-snapshot
    storageClassName: ceph-rbd
    resources:
      requests:
        cpu: 500m
        memory: 2Gi
      limits:
        memory: 2Gi
  volumeSnapshotClassName: ceph-rbd-snapclass
  deletionPolicy: Delete
kubectl -n demo apply -f elasticsearch-snapshot.yaml
kubectl -n demo wait branch/elasticsearch-snapshot --for=condition=Ready --timeout=60m
kubectl -n demo get branch elasticsearch-snapshot
kubectl -n demo get volumesnapshots
kubectl -n dev get volumesnapshots

The compatible VolumeGroupSnapshotClass is discovered separately. Individual per-volume fallback does not provide an atomic snapshot across volumes.

For cross-namespace or cross-cluster imports, the named VolumeSnapshotClass must exist wherever it is used. Cross-cluster storage must resolve handles against the same Ceph backend. See snapshot configuration.

kubectl -n demo delete branch elasticsearch-snapshot --wait=true --timeout=30m

Deletion policies

PolicyWhen the Branch is deleted
Delete (default)Remove the managed target database, cloned PVCs, action Jobs, and branch snapshot resources.
OrphanStop refresh and branch management; retain the target database and required storage and credentials, while cleaning up branch snapshot resources.

Both policies preserve the source and namespaces. Keep the controllers and storage available until finalizer cleanup finishes.

Retain a branch with Orphan

Save the following as elasticsearch-kept.yaml:

apiVersion: courier.kubedb.com/v1alpha1
kind: Branch
metadata:
  name: elasticsearch-kept
  namespace: demo
spec:
  source:
    databaseRef:
      apiGroup: kubedb.com
      kind: Elasticsearch
      name: elasticsearch-source
    namespace: demo
  target:
    namespace: dev
    name: elasticsearch-kept
    storageClassName: ceph-rbd
    resources:
      requests:
        cpu: 500m
        memory: 2Gi
      limits:
        memory: 2Gi
  deletionPolicy: Orphan
kubectl -n demo apply -f elasticsearch-kept.yaml
kubectl -n demo wait branch/elasticsearch-kept --for=condition=Ready --timeout=60m
kubectl -n demo get branch elasticsearch-kept
kubectl -n demo delete branch elasticsearch-kept --wait=true --timeout=30m
kubectl -n dev get elasticsearch elasticsearch-kept
kubectl -n dev get pvc,secrets -l app.kubernetes.io/instance=elasticsearch-kept

Verify that the retained database still serves data:

Discover the auth Secret referenced by elasticsearch-kept:

kubectl -n dev get elasticsearch elasticsearch-kept -o jsonpath='{.spec.authSecret.name}{"\n"}'
elasticsearch-kept-auth

Use the returned name below if it differs. Read its password:

kubectl -n dev get secret elasticsearch-kept-auth -o jsonpath='{.data.password}' | base64 -d

Enter this password when the client prompts.

From your terminal, open a pod shell:

kubectl -n dev exec -it elasticsearch-kept-0 -c elasticsearch -- bash

Run the following curl commands inside the pod shell. --user elastic prompts for the decoded password on each request. Requests use the database Service.

curl --silent --show-error --fail-with-body --user elastic \
  -X GET "http://elasticsearch-kept.dev.svc:9200/customers/_count?filter_path=count&pretty"
{
  "count": 5
}
curl --silent --show-error --fail-with-body --user elastic \
  -X GET "http://elasticsearch-kept.dev.svc:9200/customers/_search?size=5&sort=id:asc&filter_path=hits.hits._source&pretty"
{
  "hits": {
    "hits": [
      {"_source": {"id": 1, "full_name": "Updated Customer 1", "email": "[email protected]"}},
      {"_source": {"id": 2, "full_name": "Customer 2", "email": "[email protected]"}},
      {"_source": {"id": 3, "full_name": "Customer 3", "email": "[email protected]"}},
      {"_source": {"id": 4, "full_name": "Customer 4", "email": "[email protected]"}},
      {"_source": {"id": 5, "full_name": "Customer 5", "email": "[email protected]"}}
    ]
  }
}

Run exit to return to your terminal.

The expected values assume you completed the source update in the same-cluster tutorial. Preserve the auth Secret; the retained database still needs its credential. To explicitly remove the retained database and its durable data:

kubectl -n dev patch elasticsearch elasticsearch-kept --type=merge -p '{"spec":{"deletionPolicy":"WipeOut"}}'
kubectl -n dev delete elasticsearch elasticsearch-kept --wait=true --timeout=30m

TLS-enabled sources

Configure a TLS-enabled source following the Elasticsearch TLS guide before this example. Substitute its name in source.databaseRef.name if it differs from elasticsearch-source. Prepare cert-manager, an Issuer named branch-ca, and its signing Secret in dev on the target cluster. Courier does not install that issuer.

Set target.issuerRef when the source has enableSSL: true. Transport encryption alone does not require a user-provided issuer.

Save the following as elasticsearch-tls.yaml:

apiVersion: courier.kubedb.com/v1alpha1
kind: Branch
metadata:
  name: elasticsearch-tls
  namespace: demo
spec:
  source:
    databaseRef:
      apiGroup: kubedb.com
      kind: Elasticsearch
      name: elasticsearch-source
    namespace: demo
  target:
    namespace: dev
    name: elasticsearch-tls
    storageClassName: ceph-rbd
    resources:
      requests:
        cpu: 500m
        memory: 2Gi
      limits:
        memory: 2Gi
    issuerRef:
      apiGroup: cert-manager.io
      kind: Issuer
      name: branch-ca
  deletionPolicy: Delete
kubectl -n demo apply -f elasticsearch-tls.yaml
kubectl -n demo wait branch/elasticsearch-tls --for=condition=Ready --timeout=60m
kubectl -n demo get branch elasticsearch-tls
kubectl -n dev get certificates
kubectl -n dev get elasticsearch elasticsearch-tls -o yaml

Confirm that target certificates are ready and use branch-specific names. Use HTTPS and the issuer’s CA with curl --cacert for client verification. The plaintext commands and action connection settings from the main walkthroughs must be adapted to that TLS policy.

Use kind: ClusterIssuer when referencing a cluster-scoped issuer. A namespaced Issuer and its signing Secret must exist in the target namespace, including on another cluster.

Cleanup optional examples

kubectl -n demo delete branch elasticsearch-tls --ignore-not-found --wait=true --timeout=30m

Return to the same-cluster cleanup after all dependent Branches are gone.