Introduction
OpenShift Data Foundation (ODF), formerly known as OpenShift Container Storage (OCS), is Red Hat’s unified and software-defined storage solution for OpenShift environments. It enables persistent storage for containers, integrated backup and disaster recovery, and multicloud data management.
One of the most common deployment methods for ODF is Internal Mode, where the storage devices are hosted within the OpenShift cluster itself — ideal for small to medium-scale deployments.
Architecture Overview: Internal Mode
In Internal Mode, OpenShift Data Foundation relies on Ceph — a highly scalable storage system — and utilizes three core components:
Rook Operator
Handles deployment and lifecycle management of Ceph clusters inside Kubernetes.
Ceph Cluster (Mon, OSD, MGR, etc.)
Provides object, block, and file storage using the available storage devices on OpenShift nodes.
NooBaa
Manages object storage interfaces (S3-compatible) and acts as a data abstraction layer for multicloud object storage.
🧱 Core Storage Layers:
Object Storage Daemons (OSDs): Store actual data and replicate across nodes for redundancy.
Monitor (MON): Ensures consistency and cluster health.
Manager (MGR): Provides metrics, dashboard, and cluster management.
📦 Key Benefits of Internal Mode
No need for external storage infrastructure.
Faster to deploy and manage via OpenShift Console.
Built-in replication and self-healing mechanisms.
Ideal for lab environments, edge, or dev/test clusters.
🚀 Deployment Prerequisites
OpenShift 4.10+ cluster with minimum 3 worker nodes, each with:
At least 16 CPU cores and 64 GB RAM.
At least one unused raw block device (no partitions or file systems).
Internet connectivity or local OperatorHub mirror.
Persistent worker node roles (not shared with infra/control plane).
🔧 Steps to Deploy ODF in Internal Mode
- Install ODF Operator Go to OperatorHub in the OpenShift Console.
Search and install OpenShift Data Foundation Operator in the appropriate namespace.
- Create StorageCluster Use the ODF Console to create a new StorageCluster.
Select Internal Mode.
Choose eligible nodes and raw devices.
Validate and apply.
- Monitor Cluster Health Access the ODF dashboard from the OpenShift Console.
Verify the status of MON, OSD, and MGR components.
Monitor used and available capacity.
- Create Storage Classes Default storage classes (like ocs-storagecluster-ceph-rbd, ocs-storagecluster-cephfs) are auto-created.
Use these classes in PVCs for your applications.
🛠 Use Cases Supported
Stateful Applications: Databases (PostgreSQL, MongoDB), Kafka, ElasticSearch.
CI/CD Pipelines requiring persistent storage.
Backup and Disaster Recovery via ODF and ACM.
AI/ML Workloads needing large-scale data persistence.
📌 Best Practices
Label nodes intended for storage to prevent scheduling other workloads.
Always monitor disk health and usage via the dashboard.
Regularly test failover and recovery scenarios.
For production, consider External Mode or Multicloud Gateway for advanced scalability.
🎯 Conclusion
Deploying OpenShift Data Foundation in Internal Mode is a robust and simplified way to bring storage closer to your workloads. It ensures seamless integration with OpenShift, eliminates the need for external SAN/NAS, and supports a wide range of use cases — all while leveraging Ceph’s proven resilience.
Whether you're running apps at the edge, in dev/test, or need flexible persistent storage, ODF with Internal Mode is a solid choice.
For more info, Kindly follow: Hawkstack Technologies
Top comments (0)