How to Deploy Apache Cassandra on Vultr Kubernetes Engine

Introduction

Apache Cassandra is an open-source distributed NoSQL database designed to handle large volumes of data across multiple commodity servers. It's distributed architecture avoids single points of failure and enables horizontal scalability. Cassandra excels at write-heavy workloads and offers high write and read throughput, making it ideal for data-intensive applications. It also provides tunable consistency, accommodating varying data consistency needs.

The K8ssandra project is a collection of components, including a Kubernetes operator designed to automate the management of Apache Cassandra clusters in a Kubernetes cluster. In this article, set up a multi-node Apache Cassandra cluster in a Vultr Kubernetes Engine (VKE) cluster using the K8ssandra Operator.

Prerequisites

Before you begin:

• Deploy a Vultr Kubernetes Engine (VKE) with at least 4 nodes and 4 GB RAM per node

• Deploy a Ubuntu Server on Vultr to use as the development machine

• Access the server using SSHas a non-root user with sudo privileges

• Install and Configure Kubectl to access the cluster

• Using the Python pip package manager, install the Cassandra cqlsh CLI tool:

  console Copy

  $ pip3 install -U cqlsh
  

• Install the Helm package manager:

  console Copy

  $ snap install helm --classic
  

Install Cert-Manager

Cert-Manager is a Kubernetes operator that manages and issues TLS/SSL certificates within a cluster from trusted authorities such as Let's Encrypt. K8ssandra uses cert-manager to automate certificate management within a Cassandra clusters. This includes creating the Java keystores and truststores needed from the certificates. Follow the steps in this section to install the cert-manager resources required by the K8ssandra Operator.

1. Using Helm, add the Cert-Manager Helm repository to your local repositories.

   console Copy

   $ helm repo add jetstack https://charts.jetstack.io
   

2. Update the local Helm charts index.

   console Copy

   $ helm repo update
   

3. Install Cert-Manager to your VKE cluster.

   console Copy

   $ helm install cert-manager jetstack/cert-manager --namespace cert-manager --create-namespace --set installCRDs=true
   

4. When successful, verify that all Cert-Manager resources are available in the cluster

   console Copy

   $ kubectl get all -n cert-manager
   

   Output:

   NAME                                           READY   STATUS    RESTARTS   AGE
   pod/cert-manager-5f68c9c6dd-stmp6              1/1     Running   0          35h
   pod/cert-manager-cainjector-57d6fc9f7d-gwqr5   1/1     Running   0          35h
   pod/cert-manager-webhook-5b7ffbdc98-sq8kg      1/1     Running   0          35h
   
   NAME                           TYPE        CLUSTER-IP     EXTERNAL-IP   PORT(S)    AGE
   service/cert-manager           ClusterIP   10.102.38.47   <none>        9402/TCP   35h
   service/cert-manager-webhook   ClusterIP   10.97.255.91   <none>        443/TCP    35h
   
   NAME                                      READY   UP-TO-DATE   AVAILABLE   AGE
   deployment.apps/cert-manager              1/1     1            1           35h
   deployment.apps/cert-manager-cainjector   1/1     1            1           35h
   deployment.apps/cert-manager-webhook      1/1     1            1           35h
   
   NAME                                                 DESIRED   CURRENT   READY   AGE
   replicaset.apps/cert-manager-5f68c9c6dd              1         1         1       35h
   replicaset.apps/cert-manager-cainjector-57d6fc9f7d   1         1         1       35h
   replicaset.apps/cert-manager-webhook-5b7ffbdc98      1         1         1       35h

Install the K8ssandra Operator

1. Add the K8ssandra operator repository to your Helm sources.

   console Copy

   $ helm repo add k8ssandra https://helm.k8ssandra.io/stable
   

2. Install the K8ssandra operator in your cluster.

   console Copy

   $ helm install k8ssandra-operator k8ssandra/k8ssandra-operator -n k8ssandra-operator --create-namespace
   

3. Wait for at least 3 minutes and view the cluster deployment to verify that the K8ssandra operator is available

   console Copy

   $ kubectl -n k8ssandra-operator get deployment
   

   Your output should look like the one below:

   NAME                               READY   UP-TO-DATE   AVAILABLE   AGE
   k8ssandra-operator                 0/1     1            1           10s
   k8ssandra-operator-cass-operator   0/1     1            1           10s

4. Verify that the K8ssandra operator pods are ready and running

   console Copy

   $ kubectl get pods -n k8ssandra-operator
   

   Your output should look like the one below:

   NAME                                               READY   STATUS              RESTARTS   AGE
   k8ssandra-operator-765bcf99bf-7jfmj                0/1     ContainerCreating   0          11s
   k8ssandra-operator-cass-operator-b9cc84556-hb6jv   0/1     Running             0          11s

Set Up a Multi-node Apache Cassandra Cluster on VKE

1. Using a text editor such as nano, create a new manifest file cluster.yaml.

   console Copy

   $ nano cluster.yaml
   

2. Add the following contents to the file.

   yaml Copy

   apiVersion: k8ssandra.io/v1alpha1
   kind: K8ssandraCluster
   metadata:
     name: demo
   spec:
     cassandra:
       serverVersion: "4.0.1"
       datacenters:
         - metadata:
             name: dc1
           size: 3
           storageConfig:
             cassandraDataVolumeClaimSpec:
               storageClassName: vultr-block-storage
               accessModes:
                 - ReadWriteOnce
               resources:
                 requests:
                   storage: 10Gi
           config:
             jvmOptions:
               heapSize: 512M
         stargate:
           size: 1
           heapSize: 256M
   

   Save and close the file.

   The above configuration file defines the Cassandra cluster configuration with the following values:

   • Cassandra version: 4.0.1
   • Three cluster worker nodes.
   • The vultr-block-storage storage class with a 10 GB volume size per PVC.
   • TheCassandra node JVM heap size is 512 MB.
   • The Stargate node JVM is allocated 256 MB heap.

3. Apply the deployment to your cluster

   console Copy

   $ kubectl apply -n k8ssandra-operator -f cluster.yaml
   

4. Wait for at least 15 minutes and view the cluster pods.

   console Copy

   $ kubectl get pods -n k8ssandra-operator --watch
   

   Verify that all pods are ready and running similar to the output below:

   NAME                                               READY   STATUS    RESTARTS   AGE
   demo-dc1-default-sts-0                             0/2     Pending   0          3s
   demo-dc1-default-sts-1                             0/2     Pending   0          3s
   demo-dc1-default-sts-2                             0/2     Pending   0          3s
   k8ssandra-operator-765bcf99bf-7jfmj                1/1     Running   0          6m21s
   k8ssandra-operator-cass-operator-b9cc84556-hb6jv   1/1     Running   0          6m21s

   When all Cassandra database pods are ready, the Stargate Pod creation is initiated. Stargate provides a data gateway with REST, GraphQL, and Document APIs in front of the Cassandra database. The name of the Stargate Pod shoudl be similar to: demo-dc1-default-stargate-deployment-597b876d8f-559pt.

Verify the Linked Vultr Block Storage PVCs For Cassandra Cluster Persistence

The Kubernetes StatefulSet controller creates the Cassandra cluster pods which in turn is created by the K8ssandra operator during cluster deployment. StatefulSet is the key to data persistence of each node in the Cassandra cluster and there is one StatefulSet in each cluster node. Verify the available cluster PVCs to improve the Cassandra cluster persistence.

1. Verify that the StatefulSets are available and ready in your cluster

   console Copy

   $ kubectl get statefulset -n k8ssandra-operator
   

   Output:

   NAME                   READY   AGE
   demo-dc1-default-sts   3/3     15m

2. Verify the available Storage class.

   console Copy

   $ kubectl get sc vultr-block-storage
   

   Output:

   NAME                  PROVISIONER           RECLAIMPOLICY   VOLUMEBINDINGMODE   ALLOWVOLUMEEXPANSION   AGE
   vultr-block-storage   block.csi.vultr.com   Delete          Immediate           true                   36m

   As displayed in the above output, Vultr offers both HDD and NVME block storage technologies.The Vultr Container Storage Interface (CSI) to connect your VKE cluster to Vultr block storage and deploy the high performance NVME class. It is automatically deployed by the managed control plane in the VKE cluster.

3. To verify the deployed Vultr Block Storage volumes attached to your VKE cluster, view all cluster PVCs.

   console Copy

   $ kubectl get pvc --all-namespaces
   

   In addition, open the Vultr Customer Postal and navigate to the Linked Resources tab in your VKE cluster control panel:

   

   You can also verify the linked volumes by navigating to your Vultr Account Block Storage Page.

   

Create a Kubernetes Service to Access the Cassandra cluster

1. Create a new service resource file service.yaml.

   console Copy

   $ nano service.yaml
   

2. Add the following contents to the file.

   yaml Copy

   apiVersion: v1
   kind: Service
   metadata:
     name: cassandra
     labels:
       app: cassandra
   spec:
     type: LoadBalancer
     externalTrafficPolicy: Local
     ports:
       - port: 9042
     selector:
       statefulset.kubernetes.io/pod-name: demo-dc1-default-sts-1
   

   Save and close the file.

   The above configuration defines a Kubernetes service with a LoadBalancer type to access Cassandra cluster using port 9042.

3. Apply the service to your cluster.

   console Copy

   $ kubectl apply -n k8ssandra-operator -f service.yaml
   

4. Wait for at least 5 minutes to deploy the cluster Loadbalancer resource and view the Cassandra service

   console Copy

   $ kubectl get svc/cassandra -n k8ssandra-operator
   

   Verify the IP Address in your External-IP value to use for accessing the cluster

   NAME        TYPE           CLUSTER-IP      EXTERNAL-IP   PORT(S)          AGE
   cassandra   LoadBalancer   10.110.178.86   <pending>     9042:32313/TCP   3s

Test the Apache Cassandra Cluster

cqlsh is a command-line interface that allows users to connect to a Cassandra cluster. Follow the steps in this section to execute CQL (Cassandra Query Language) statements, and perform various database operations, such as creating, modifying, and querying data.

1. Export your Cassandra service Load Balancer IP to the CASS_IP variable.

   console Copy

   $ CASS_IP=$(kubectl get svc cassandra -n k8ssandra-operator -o jsonpath="{.status.loadBalancer.ingress[*].ip}")
   

   View the variable value.

   console Copy

   $ echo $CASS_IP
   

2. Export the cluster access username to the CASS_USERNAME variable.

   console Copy

   $ CASS_USERNAME=$(kubectl get secret demo-superuser -n k8ssandra-operator -o=jsonpath='{.data.username}' | base64 --decode)
   

   View the username value.

   console Copy

   $ echo $CASS_USERNAME
   

3. Export the cluster the password to the CASS_PASSWORD variable.

   console Copy

   $ CASS_PASSWORD=$(kubectl get secret demo-superuser -n k8ssandra-operator -o=jsonpath='{.data.password}' | base64 --decode)
   

   View the password value.

   console Copy

   $ echo $CASS_PASSWORD
   

4. Using cqlsh, log in to the Cassandra cluster using your variable values.

   console Copy

   $ cqlsh -u $CASS_USERNAME -p $CASS_PASSWORD $CASS_IP 9042
   

5. Create a new keyspace demo in the Cassandra database.

   sql Copy

   > CREATE KEYSPACE demo WITH replication = {'class': 'SimpleStrategy', 'replication_factor': 3};
   

6. Create a new table users in the demo keyspace.

   sql Copy

   > CREATE TABLE demo.users (id text primary key, name text, country text);
   

7. Add data to the users table.

   sql Copy

   > INSERT INTO demo.users (id, name, country) values ('42', 'John Doe', 'UK');
   > INSERT INTO demo.users (id, name, country) values ('43', 'Joe Smith', 'US');
   

8. Query the table data to view the stored values.

   sql Copy

   > SELECT * FROM demo.users;
   

   Output:

   id | country | name
   ----+---------+-----------
   43 |      US | Joe Smith
   42 |      UK | John Doe
   
   (2 rows)

Conclusion

You have deployed Apache Cassandra in a Vultr Kubernetes Engine (VKE) cluster using the open-source K8ssandra cluster. In addition, you set up Vultr Block Storage for data persistence and accessed the Cassandra cluster using the cqlsh CLI tool. For more information and Cassandra operations, visit the official documentation.

More Information

• How to Implement Apache Cassandra Driver for Python
• Vultr Block Storage
• k8ssandra documentation