The project is currently alpha. While no breaking API changes are currently planned, we reserve the right to address bugs and change the API before the project is declared stable.
This operator runs a Zookeeper 3.5 cluster, and uses Zookeeper dynamic reconfiguration to handle node membership.
The operator itself is built with the Operator framework.
- Access to a Kubernetes v1.9.0+ cluster
Note: if you are running on Google Kubernetes Engine (GKE), please check this first.
Register the ZookeeperCluster custom resource definition (CRD).
$ kubectl create -f deploy/crd.yaml
You can choose to enable Zookeeper operator for all namespaces or just for the a specific namespace. The example is using the default namespace, but feel free to edit the Yaml files and use a different namespace.
Create the operator role and role binding.
// default namespace
$ kubectl create -f deploy/default_ns/rbac.yaml
// all namespaces
$ kubectl create -f deploy/all_ns/rbac.yaml
Deploy the Zookeeper operator.
// default namespace
$ kubectl create -f deploy/default_ns/operator.yaml
// all namespaces
$ kubectl create -f deploy/all_ns/operator.yaml
Verify that the Zookeeper operator is running.
$ kubectl get deploy
NAME DESIRED CURRENT UP-TO-DATE AVAILABLE AGE
zookeeper-operator 1 1 1 1 12m
Create a Yaml file called zk.yaml with the following content to install a 3-node Zookeeper cluster.
apiVersion: "zookeeper.pravega.io/v1beta1"
kind: "ZookeeperCluster"
metadata:
name: "example"
spec:
size: 3$ kubectl create -f zk.yaml
Verify that the cluster instances and its components are running.
$ kubectl get zk
NAME AGE
example 15s
$ kubectl get all -l app=example
NAME DESIRED CURRENT AGE
statefulsets/example 3 3 2m
NAME READY STATUS RESTARTS AGE
po/example-0 1/1 Running 0 2m
po/example-1 1/1 Running 0 1m
po/example-2 1/1 Running 0 1m
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
svc/example-client ClusterIP 10.31.243.173 <none> 2181/TCP 2m
svc/example-headless ClusterIP None <none> 2888/TCP,3888/TCP 2m
$ kubectl delete -f zk.yaml
Note that the Zookeeper clusters managed by the Zookeeper operator will NOT be deleted even if the operator is uninstalled.
To delete all clusters, delete all cluster CR objects before uninstalling the operator.
$ kubectl delete -f deploy/default_ns
// or, depending on how you deployed it
$ kubectl delete -f deploy/all_ns
Requirements:
- Go 1.10+
- Operator SDK
Use the operator-sdk command to build the Zookeeper operator image.
$ operator-sdk build pravega/zookeeper-operator
The Zookeeper operator image will be available in your Docker environment.
$ docker images pravega/zookeeper-operator
REPOSITORY TAG IMAGE ID CREATED SIZE
pravega/zookeeper-operator latest 7e735b292e14 59 seconds ago 37.2MB
Optionally push it to a Docker registry.
docker tag pravega/zookeeper-operator [REGISTRY_HOST]:[REGISTRY_PORT]/pravega/zookeeper-operator
docker push [REGISTRY_HOST]:[REGISTRY_PORT]/pravega/zookeeper-operator
where:
[REGISTRY_HOST]is your registry host or IP (e.g.registry.example.com)[REGISTRY_PORT]is your registry port (e.g.5000)
For debugging and development you might want to access the Zookeeper cluster directly. For example, if you created the cluster with name example in the default namespace you can forward the Zookeeper port from any of the pods (e.g. example-0) as follows:
$ kubectl port-forward -n default example-0 2181:2181
You can run the operator locally to help with development, testing, and debugging tasks.
The following command will run the operator locally with the default Kubernetes config file present at $HOME/.kube/config. Use the --kubeconfig flag to provide a different path.
$ operator-sdk up local
The Operator requires elevated privileges in order to watch for the custom resources.
According to Google Container Engine docs:
Ensure the creation of RoleBinding as it grants all the permissions included in the role that we want to create. Because of the way Container Engine checks permissions when we create a Role or ClusterRole.
An example workaround is to create a RoleBinding that gives your Google identity a cluster-admin role before attempting to create additional Role or ClusterRole permissions.
This is a known issue in the Beta release of Role-Based Access Control in Kubernetes and Container Engine version 1.6.
On GKE, the following command must be run before installing the operator, replacing the user with your own details.
$ kubectl create clusterrolebinding your-user-cluster-admin-binding --clusterrole=cluster-admin --user=your.google.cloud.email@example.org