Kubernetes

GreptimeDB Operator

By utilizing the Operator pattern, GreptimeDB Operator can effectively manage GreptimeDB clusters on Kubernetes. This operator abstracts the model of maintaining a highly available GreptimeDB cluster.

1. Create a test Kubernetes cluster

You can use kind to create your own test Kubernetes cluster:

  1. kind create cluster

2. Use Helm to install GreptimeDB Operator

Make sure you have already installed Helm. Use the following commands to install greptimedb-operator in the default namespace:

  1. helm repo add greptime https://greptimeteam.github.io/helm-charts/
  1. helm repo update
  1. helm install greptimedb-operator greptime/greptimedb-operator -n default

The maintained Helm charts are in helm-charts.

3. Create your own GreptimeDB cluster

Create an etcd cluster for GreptimeDB:

  1. helm install etcd oci://registry-1.docker.io/bitnamicharts/etcd \
  2.   --set replicaCount=3 \
  3.   --set auth.rbac.create=false \
  4.   --set auth.rbac.token.enabled=false \
  5.   -n default

Create a GreptimeDB cluster which uses the etcd cluster created at previous step:

  1. helm install mycluster greptime/greptimedb-cluster -n default

Or, if you already have an etcd cluster, you can use etcdEndpoints to use your etcd cluster:

  1. helm install mycluster greptime/greptimedb-cluster \
  2.   --set etcdEndpoints=<your-etcd-cluster-endpoints> \
  3.   -n default

After the installation, you can use kubectl port-forward to forward the MySQL protocol port of the GreptimeDB cluster:

  1. # You can use the MySQL client to connect the cluster, for example: 'mysql -h 127.0.0.1 -P 4002'.
  2. kubectl port-forward svc/mycluster-frontend 4002:4002 > connections.out &
  4. # You can use the PostgreSQL client to connect the cluster, for example: 'psql -h 127.0.0.1 -p 4003 -d public'.
  5. kubectl port-forward svc/mycluster-frontend 4003:4003 > connections.out &

Then you can use MySQL client to connect to the cluster.

4. Destroy GreptimeDB cluster

You can use the following commands to uninstall operator and cluster:

  1. # Uninstall the cluster.
  2. helm uninstall mycluster -n default
  1. # Uninstall etcd.
  2. helm uninstall etcd -n default
  1. # Uninstall the operator.
  2. helm uninstall greptimedb-operator -n default
  1. # Delete crds.
  2. kubectl delete crds greptimedbclusters.greptime.io