How to Provision Persistent Volume Claims on Vultr Kubernetes Engine

Vultr offers persistent storage options for Kubernetes through both Vultr Block Storage (VBS) and the Vultr File System (VFS). These storage solutions are integrated with the Vultr Kubernetes Engine (VKE) to support a wide range of application needs—from high-performance single-node volumes to shared, multi-node file systems. VKE includes a managed Kubernetes control plane that automatically installs the Vultr Container Storage Interface (CSI), enabling seamless access to these storage types and allowing your applications to scale with ease.

In this guide, you will provision persistent volumes using both Vultr Block Storage (VBS) and Vultr File System (VFS), and learn how to update the configuration to perform PVC expansion where supported.

Prerequisites

Before you begin:

• Deploy a Vultr Ubuntu server
• Deploy a Vultr Kubernetes Engine (VKE) cluster with at least 2 nodes.
• Access the workstation using SSH as a non-root user with sudo privileges.
• Install and configure Kubectl to access the cluster
• Vultr Container Storage Interface (CSI) Driver v0.14.0 or higher.
• Kubernetes Config (KUBECONFIG) for cluster management.

Understanding Storage Classes

Vultr provides two types of persistent storage technologies for Kubernetes workloads, available in most global regions: Vultr Block Storage and Vultr File System (VFS). These options support a variety of performance, capacity, and access mode requirements.

Vultr Block Storage (VBS)

Vultr Block Storage supports ReadWriteOnce (RWO) access mode and is ideal for high-performance or large-capacity workloads where each volume is attached to a single node.

• HDD Block Storage – A cost-effective solution using traditional spinning hard drives. Suitable for workloads that require large capacity over speed.

  • CSI Storage Class: vultr-block-storage-hdd
  • Minimum Volume Size: 40 GiB
  • Maximum Volume Size: 40 TiB
  • Technology: Rotational hard disk drive (HDD)
  • Availability: Most Vultr regions
  • Key Benefit: Affordable storage with support for very large volumes

• NVMe Block Storage – A high-performance option backed by solid-state NVMe drives. Best for latency-sensitive or IOPS-intensive applications.

  • CSI Storage Class: vultr-block-storage
  • Minimum Volume Size: 10 GiB
  • Maximum Volume Size: 10 TiB
  • Technology: Solid-state NVMe
  • Availability: Many Vultr regions
  • Key Benefit: Fastest I/O performance for demanding workloads

Vultr File System (VFS)

The Vultr File System provides ReadWriteMany (RWX) access, allowing multiple pods across different nodes to read and write to the same volume simultaneously. It’s ideal for shared data use cases such as content management systems, collaborative apps, and distributed workloads.

• CSI Storage Class: vultr-vfs-storage
• Minimum Volume Size: 10 GiB
• Maximum Volume Size: 10 TiB
• Technology: Network-attached file system
• Availability: Selected Vultr regions
• Key Benefit: Shared storage for multi-pod and multi-node access (RWX)

To check availability in your region, use the /v2/regions API endpoint.

Look for the following flags in the response:

• block_storage_storage_opt: Indicates HDD Block Storage availability
• block_storage_high_perf: Indicates NVMe Block Storage availability

Some locations support both storage classes. If NVMe block storage is available in a location, our CSI uses that class by default.

Create Persistent Volume Claims (PVCs) with Vultr Block Storage (VBS)

In this section, you are to create a PVC using a YAML configuration file with ReadWriteOnce (RWO) access mode.

1. Create a configuration file.

   console Copy

   $ nano create_pvc.yaml
   

2. Copy and paste the below given configuration.

   yaml Copy

   apiVersion: v1
   kind: PersistentVolumeClaim
   metadata:
     name: csi-pvc
   spec:
     accessModes:
       - ReadWriteOnce
     resources:
       requests:
         storage: 10Gi
     storageClassName: vultr-block-storage
   

   Save and exit the file.

   In the above configuration, you are requesting an NVMe block storage volume with a size of 10Gi using the vultr-block-storage storage class. This PVC uses the ReadWriteOnce access mode, meaning it can be mounted as read-write by a single node.

   • Supported Vultr Block Storage types and minimum size requirements:

     • NVMe Block Storage

       • storageClassName: vultr-block-storage
       • Minimum size: 10Gi

     • HDD Block Storage

       • storageClassName: vultr-block-storage-hdd
       • Minimum size: 40Gi

     To use HDD block storage, update the storageClassName to vultr-block-storage-hdd and set the storage request to at least 40Gi in the above configuration.

3. Apply the configuration file.

   console Copy

   $ sudo kubectl apply -f create_pvc.yaml
   

   Output

   persistentvolumeclaim/csi-pvc created

4. Verify that a new Vultr Block Storage is provisioned.

   console Copy

   $ sudo kubectl get pvc
   

   Output

   NAME      STATUS   VOLUME                 CAPACITY   ACCESS MODES   STORAGECLASS          VOLUMEATTRIBUTESCLASS   AGE
   csi-pvc   Bound    pvc-36bbc0eb476040a9   10Gi       RWO            vultr-block-storage   <unset>                 56s

5. You can verify the same by navigating to Cloud Storage then Block Storage on your Vultr Customer Portal.

   

PVC Expansion

In this section, you are to expand the storage capacity of an existing VBS attached to your instance.

1. Change the configuration file.

   console Copy

   $ nano create_pvc.yaml
   

2. Increase the storage capacity.

   yaml Copy

   apiVersion: v1
   kind: PersistentVolumeClaim
   metadata:
     name: csi-pvc
   spec:
     accessModes:
       - ReadWriteOnce
     resources:
       requests:
         storage: 30Gi
     storageClassName: vultr-block-storage
   

   Save and exit the file.

   In the above configuration, you increased the storage capacity for an existing NVMe block storage to 30 GB from the original 10 GB.

3. Apple the configuration changes.

   console Copy

   $ sudo kubectl apply -f create_pvc.yaml
   

   Output

   persistentvolumeclaim/csi-pvc configured

4. Verify the change.

   console Copy

   $ sudo kubectl get pvc
   

   Output

   NAME      STATUS   VOLUME                 CAPACITY   ACCESS MODES   STORAGECLASS          VOLUMEATTRIBUTESCLASS   AGE
   csi-pvc   Bound    pvc-36bbc0eb476040a9   30Gi       RWO            vultr-block-storage   <unset>                 5m1s

5. You can verify the same by navigating to Cloud Storage then Block Storage on your Vultr portal.

   

6. It is not advised to reduce the storage capacity of the VBS as the action may result in significant data loss.

Create Persistent Volume Claims (PVCs) with Vultr File System (VFS)

Vultr File System volumes can be used as Persistent Volumes in Kubernetes using the vultr-vfs-storage storage class.

1. Create a new manifest for PVC resource.

   yaml Copy

   apiVersion: v1
   kind: PersistentVolumeClaim
   metadata:
     name: test-rwx
   spec:
     accessModes:
       - ReadWriteMany
     resources:
       requests:
         storage: 10Gi
     storageClassName: vultr-vfs-storage
   

   In this example, the PVC resource test-rwx is created with a capacity of 10Gi and access mode ReadWriteMany. The storage class vultr-vfs-storage is used to provision the VFS volume.

2. Apply the PVC manifest.

   console Copy

   $ kubectl apply -f pvc.yaml
   

   Output:

   persistentvolumeclaim/test-rwx created

3. List the PVC resource.

   console Copy

   $ kubectl get pvc test-rwx
   

   The PVC resource test-rwx should be listed with the status Bound.

   Output:

   NAME          STATUS   VOLUME                 CAPACITY   ACCESS MODES   STORAGECLASS        VOLUMEATTRIBUTESCLASS   AGE
   test-rwx      Bound    pvc-604cf0c549ad41e3   10Gi       RWX            vultr-vfs-storage   <unset>                 10m

Troubleshooting

If you are unable to reproduce the same behaviour as noted above, you can try troubleshooting to get more information.

• Describe the PVC resource.

  console Copy

  $ kubectl describe pvc test-rwx
  

  You can review the events for any issues related to provisioning of the VFS volume.

• List all storage classes.

  console Copy

  $ kubectl get storageclass
  

  The list must contain vultr-vfs-storage in order to provision a VFS volume.

• Ensure sufficient quota allocation.

• Monitor cluster events for storage issues.

Conclusion

In this guide, you learned about the different persistent storage types and classes available globally on Vultr, including both Vultr Block Storage (VBS) and Vultr File System (VFS). You provisioned an HDD block storage volume and also explored how to expand the storage of an existing volume attached to a Kubernetes cluster.