Skip to main content
Version: nightly

Deploy GreptimeDB Operator

Overview​

The GreptimeDB Kubernetes Operator simplifies deploying GreptimeDB on both private and public cloud infrastructures. This guide walks you through installing the latest stable version of the GreptimeDB Operator on a Kubernetes cluster. The Operator leverages a Custom Resource Definition (CRD) to represent GreptimeDB as a Kubernetes object.

Prerequisites​

  • Helm (Use the Version appropriate for your Kubernetes API version)

Install the GreptimeDB Operator using Helm Charts​

GreptimeDB provides a Helm-compatible repository for easy deployment. Follow these steps to install the Operator using Helm:

Add the GreptimeDB Operator repository​

First, add the GreptimeDB Operator Helm repository:

helm repo add greptime https://greptimeteam.github.io/helm-charts/

Validate the repository by searching for the Operator chart:

helm search repo greptimedb-operator

You should see output similar to this:

NAME                        	CHART VERSION	APP VERSION   	DESCRIPTION
greptime/greptimedb-operator	0.2.3        	0.1.0-alpha.29	The greptimedb-operator Helm chart for Kubernetes.

Install the Operator​

To install the Operator, run the following helm install command. This command also creates a dedicated namespace greptimedb-admin for the installation. It's recommended to use a dedicated namespace for the Operator:

helm install \
  operator greptime/greptimedb-operator \
  --create-namespace \
  -n greptimedb-admin

Verify the CRD installation​

Check the contents of the greptimedb-admin namespace to confirm that all Custom Resource Definitions (CRDs) have been installed correctly:

kubectl get crds -n greptimedb-admin

You should see output similar to this:

NAME                                CREATED AT
greptimedbclusters.greptime.io      2024-09-09T07:54:07Z
greptimedbstandalones.greptime.io   2024-09-09T07:54:07Z

Verify the Operator installation​

After installation, check the contents of the greptimedb-admin namespace to confirm that all pods are running correctly:

kubectl get pods -n greptimedb-admin

You should see output similar to this:

NAME                                            READY   STATUS    RESTARTS   AGE
operator-greptimedb-operator-7d58cb8f7c-jz46g   1/1     Running   0          26s