KubeDB

By AppsCode

CONTACT SALES

New to KubeDB? Please start here.

Horizontal Scale Weaviate Cluster

This guide will show you how to use the KubeDB Ops Manager to scale the number of nodes of a Weaviate cluster up and down.

Before You Begin

  • At first, you need to have a Kubernetes cluster, and the kubectl command-line tool must be configured to communicate with your cluster.

  • Install KubeDB in your cluster following the steps here.

  • You should be familiar with the following KubeDB concepts:

To keep things isolated, this tutorial uses a separate namespace called demo throughout this tutorial.

$ kubectl create ns demo
namespace/demo created

Note: YAML files used in this tutorial are stored in docs/examples/weaviate/scaling/horizontal-scaling folder in GitHub repository kubedb/docs.

Deploy Weaviate

In this section, we are going to deploy a Weaviate cluster with 3 nodes.

apiVersion: kubedb.com/v1alpha2
kind: Weaviate
metadata:
  name: weaviate-sample
  namespace: demo
spec:
  version: 1.33.1
  replicas: 3
  storageType: Durable
  storage:
    storageClassName: longhorn
    accessModes:
      - ReadWriteOnce
    resources:
      requests:
        storage: 1Gi
  deletionPolicy: WipeOut

Let’s create the Weaviate CR and wait for it to become Ready:

$ kubectl get weaviate -n demo
NAME              TYPE                  VERSION   STATUS   AGE
weaviate-sample   kubedb.com/v1alpha2   1.33.1    Ready    5m

$ kubectl get pods -n demo -l app.kubernetes.io/instance=weaviate-sample
NAME                READY   STATUS    RESTARTS   AGE
weaviate-sample-0   1/1     Running   0          5m
weaviate-sample-1   1/1     Running   0          5m
weaviate-sample-2   1/1     Running   0          5m

Scale Up

Here, we are going to scale up the cluster from 3 nodes to 5 nodes.

apiVersion: ops.kubedb.com/v1alpha1
kind: WeaviateOpsRequest
metadata:
  name: weaviate-scale-up
  namespace: demo
spec:
  type: HorizontalScaling
  databaseRef:
    name: weaviate-sample
  horizontalScaling:
    node: 5
  • spec.type specifies that this is a HorizontalScaling operation.
  • spec.horizontalScaling.node specifies the desired number of nodes after scaling.

Let’s create the WeaviateOpsRequest CR:

$ kubectl create -f https://github.com/kubedb/docs/raw/v2026.7.10/docs/examples/weaviate/scaling/horizontal-scaling/scale-up.yaml
weaviateopsrequest.ops.kubedb.com/weaviate-scale-up created

The Ops Manager adds the new nodes, waits for them to join and sync the schema, and rebalances the shard replicas.

$ kubectl get weaviateopsrequest -n demo weaviate-scale-up
NAME                TYPE                STATUS       AGE
weaviate-scale-up   HorizontalScaling   Successful   3m

$ kubectl get weaviateopsrequest -n demo weaviate-scale-up -o yaml
...
status:
  conditions:
  - message: Weaviate ops-request has started horizontal scaling
    reason: HorizontalScaling
    status: "True"
    type: HorizontalScaling
  - message: patch petset; ConditionStatus:True
    status: "True"
    type: PatchPetset
  - message: is node in cluster; ConditionStatus:True
    status: "True"
    type: IsNodeInCluster
  - message: successfully restarted new Weaviate pods for schema sync
    reason: RestartNodes
    status: "True"
    type: RestartNodes
  - message: successfully rebalanced Weaviate shard replicas
    reason: RebalanceShards
    status: "True"
    type: RebalanceShards
  - message: Successfully scaled up nodes
    reason: HorizontalScaleUp
    status: "True"
    type: HorizontalScaleUp
  - message: successfully reconciled Weaviate with new replica count
    reason: UpdatePetSets
    status: "True"
    type: UpdatePetSets
  - message: Horizontal scaling completed
    reason: Successful
    status: "True"
    type: Successful
  observedGeneration: 1
  phase: Successful

Verify the new node count:

$ kubectl get pods -n demo -l app.kubernetes.io/instance=weaviate-sample
NAME                READY   STATUS    RESTARTS   AGE
weaviate-sample-0   1/1     Running   0          3m51s
weaviate-sample-1   1/1     Running   0          3m11s
weaviate-sample-2   1/1     Running   0          2m31s
weaviate-sample-3   1/1     Running   0          56s
weaviate-sample-4   1/1     Running   0          44s

$ kubectl get weaviate -n demo weaviate-sample -o jsonpath='{.spec.replicas}'
5

Scale Down

Now, let’s scale the cluster back down from 5 nodes to 2 nodes.

apiVersion: ops.kubedb.com/v1alpha1
kind: WeaviateOpsRequest
metadata:
  name: weaviate-scale-down
  namespace: demo
spec:
  type: HorizontalScaling
  databaseRef:
    name: weaviate-sample
  horizontalScaling:
    node: 2

Let’s create the WeaviateOpsRequest CR:

$ kubectl create -f https://github.com/kubedb/docs/raw/v2026.7.10/docs/examples/weaviate/scaling/horizontal-scaling/scale-down.yaml
weaviateopsrequest.ops.kubedb.com/weaviate-scale-down created

The Ops Manager moves the shards off the nodes that are going away before removing them.

$ kubectl get weaviateopsrequest -n demo weaviate-scale-down
NAME                  TYPE                STATUS       AGE
weaviate-scale-down   HorizontalScaling   Successful   3m

$ kubectl get weaviateopsrequest -n demo weaviate-scale-down -o yaml
...
status:
  conditions:
  - message: Weaviate ops-request has started horizontal scaling
    reason: HorizontalScaling
    status: "True"
    type: HorizontalScaling
  - message: move shards; ConditionStatus:True; PodName:weaviate-sample-4
    status: "True"
    type: MoveShards--weaviate-sample-4
  - message: delete pvc; ConditionStatus:True; PodName:weaviate-sample-4
    status: "True"
    type: DeletePvc--weaviate-sample-4
  - message: move shards; ConditionStatus:True; PodName:weaviate-sample-3
    status: "True"
    type: MoveShards--weaviate-sample-3
  - message: delete pvc; ConditionStatus:True; PodName:weaviate-sample-3
    status: "True"
    type: DeletePvc--weaviate-sample-3
  - message: move shards; ConditionStatus:True; PodName:weaviate-sample-2
    status: "True"
    type: MoveShards--weaviate-sample-2
  - message: delete pvc; ConditionStatus:True; PodName:weaviate-sample-2
    status: "True"
    type: DeletePvc--weaviate-sample-2
  - message: Successfully scaled down nodes
    reason: HorizontalScaleDown
    status: "True"
    type: HorizontalScaleDown
  - message: successfully reconciled Weaviate with new replica count
    reason: UpdatePetSets
    status: "True"
    type: UpdatePetSets
  - message: Horizontal scaling completed
    reason: Successful
    status: "True"
    type: Successful
  observedGeneration: 1
  phase: Successful

Verify the node count again:

$ kubectl get pods -n demo -l app.kubernetes.io/instance=weaviate-sample
NAME                READY   STATUS    RESTARTS   AGE
weaviate-sample-0   1/1     Running   0          7m27s
weaviate-sample-1   1/1     Running   0          6m47s

$ kubectl get weaviate -n demo weaviate-sample -o jsonpath='{.spec.replicas}'
2

The cluster has been scaled horizontally — first up to 5 nodes, then back down to 2.

Next Steps

Cleaning Up

To clean up the Kubernetes resources created by this tutorial, run:

$ kubectl delete weaviateopsrequest -n demo weaviate-scale-up weaviate-scale-down
$ kubectl delete weaviate -n demo weaviate-sample
$ kubectl delete ns demo

What's on this Page

Search
Improve This Page