Storage

Hello and welcome to this article on OpenShift. This guide explains how persistent storage works in OpenShift, how it is provisioned, and how applications consume it. Containers are designed to be ephemeral: they are created to process data and are removed when that processing completes. By default, data written inside a container is lost when the container is destroyed. To persist data beyond a container’s lifetime, attach persistent volumes to containers at creation time. Data written to those volumes is retained independently of the container lifecycle. OpenShift leverages the Kubernetes Persistent Volume (PV) framework to provide cluster storage. Storage can be provisioned by many plugins and providers, including Local, iSCSI, Fibre Channel, NFS, GlusterFS, Ceph RBD, OpenStack Cinder, AWS Elastic Block Store, Azure Disk/File, and VMware vSphere.

Storage exposed to the OpenShift cluster appears as a pool of PersistentVolume (PV) resources. Projects (namespaces) request storage from that pool by creating PersistentVolumeClaims (PVCs). When a PVC is bound to a PV, the claim can be mounted into pods so containers can read and write persistent data.

A stylized red 3D diagram showing Kubernetes persistent volumes: large rectangular PV blocks with floating red cube pointers labeled PVC — 10GB, PVC — 100GB, and PVC — 20GB. Stacked cylinder storage icons and the label "PV Resources — 1TB" appear below.

From the OpenShift web console you can create storage using the Create Storage button, provide a name for the claim, choose an access mode, and specify the size of the claim. Alternatively, create PVCs via YAML and apply them with the CLI (oc apply -f pvc.yaml). Access modes define how a PV can be mounted into nodes and pods:

Access Mode	What it means	Typical storage types
ReadWriteOnce (RWO)	Volume can be mounted read-write by a single node at a time.	Block-backed volumes such as AWS EBS, Azure Disk
ReadWriteMany (RWX)	Volume can be mounted read-write by multiple nodes (shared).	NFS, distributed file systems (GlusterFS, CephFS)
ReadOnlyMany (ROX)	Volume can be mounted read-only by multiple nodes.	Shared file systems for read-only access

Select the appropriate access mode and size, then create the PersistentVolumeClaim. The web UI exposes these options in a simple form:

Once a PVC exists and is bound, attach it to a pod by adding a volume that references the claim in your DeploymentConfig (OpenShift) or Deployment/Pod manifest and mounting it into the container. The minimal example below shows:

A PVC requesting 5 Gi of storage.
A DeploymentConfig that mounts the PVC at /data inside the container.

apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  name: my-pvc
spec:
  accessModes:
    - ReadWriteOnce
  resources:
    requests:
      storage: 5Gi
  storageClassName: standard

apiVersion: apps.openshift.io/v1
kind: DeploymentConfig
metadata:
  name: my-app
spec:
  replicas: 1
  selector:
    app: my-app
  template:
    metadata:
      labels:
        app: my-app
    spec:
      containers:
        - name: my-app
          image: nginx:1.21
          volumeMounts:
            - name: app-storage
              mountPath: /data
      volumes:
        - name: app-storage
          persistentVolumeClaim:
            claimName: my-pvc
  strategy:
    type: Rolling

Note that the storageClassName in your PVC controls which StorageClass (and thus which provisioner) satisfies the claim. Many OpenShift clusters support dynamic provisioning: PVCs are automatically bound to dynamically created PVs using the selected StorageClass.

Access modes are an important constraint when designing storage for your application. Use RWO for single-node writable volumes (common with block storage), RWX when multiple nodes must write concurrently (NFS or distributed filesystems), and ROX when read-only sharing is sufficient.

Be careful with deletion semantics: deleting a PVC can also delete the underlying PV depending on the StorageClass reclaim policy (e.g., Delete vs Retain). Verify reclaim policies before removing claims to avoid accidental data loss.

Common CLI and troubleshooting commands:

Create or update resources: oc apply -f <file.yaml>
List claims and volumes: oc get pvc, oc get pv
Inspect a claim or volume: oc describe pvc <name>, oc describe pv <name>
Check StorageClasses: oc get storageclass, oc describe storageclass <name>

Introduction

Pre-Requisite – Docker and Kubernetes

Getting Started with OpenShift

OpenShift Concepts – Projects and Users

Concepts – Builds and Deployments

Networks, Services, Routes and Scaling

Storage, Templates and Catalog

OpenShift Security

Conclusion and APPENDIX

Watch Video