Setting up NFS storage
If you plan to use NFS for persistent storage, you must set up your NFS storage before you install Cloud Pak for Data.
It is highly recommended that your NFS storage is served by an enterprise class storage system that ensures availability with a sufficiently high throughput and reduced latency. It is recommended that the storage is in close proximity to the cluster. For example, the storage should be in the same network as the cluster. For details, see Disk requirements.
Ensure that the following statements are true on all the nodes you plan to run Cloud Pak for Data on:
- All of the nodes in the cluster have access to mount the NFS server.
- All of the nodes in the cluster have read/write access to the NFS server.
- Containerized processes have read/write access to the NFS server.Important: Containerized processes create files that are owned by various UIDs. (In Cloud Pak for Data, most services use long UIDs between 1000320900 and 1000361000, as an example. UID range is decided by OpenShift® and project namespaces.) Db2® reserves a specific UID range because it uses a custom security context constraint (SCC). For more information, see Service UIDs. If you restrict access to the NFS served to specific UIDs, you might encounter errors when installing or running Cloud Pak for Data.
Before you begin
Determine which services you will install on IBM® Cloud Pak for Data.
At a minimum, you must configure dynamic provisioning for NFS storage.
- DataStage®
- Db2
- Db2 Big SQL
- Db2 Warehouse
- OpenPages®
- Watson™ Knowledge Catalog
- Data Virtualization
In addition, if you are running SELinux on your worker nodes, you might need to create an SELinux policy module on your worker nodes.
Use the following information to ensure that you complete the appropriate tasks for setting up your NFS storage:
| Options | What to do |
|---|---|
| I am installing a service that has a dependency on Db2U | |
| I am not installing any services that have a dependency on Db2U |