IBM / cloud-pak-deployer

Configuration-based installation of OpenShift and Cloud Pak for Data/Integration/Watson AIOps on various private and public cloud infrastructure providers. Deployment attempts to achieve the end-state defined in the configuration. If something fails along the way, you only need to restart the process to continue the deployment.
https://ibm.github.io/cloud-pak-deployer/
Apache License 2.0
140 stars 69 forks source link

Change namespace for NFS from cp-nfs to nfs-provisioner #800

Closed Alan111S closed 1 month ago

Alan111S commented 1 month ago

Not exactly a bug but a setting which can cause an issue:-

The namespace name used for NFS is cp-nfs

If a CPD administrator later runs cpd-cli manage setup-nfs-provisioner commands (e.g. https://www.ibm.com/docs/en/cloud-paks/cp-data/5.0.x?topic=storage-setting-up-dynamic-provisioning), the documentation suggests using the namespace name nfs-provisioner.

Quote: "Set PROJECT_NFS_PROVISIONER environment variable to the project (namespace) where you want to deploy the Kubernetes NFS-Client Provisioner provisioner. The recommended project is nfs-provisioner; however, you can specify any project."

Since a deployer user is unlikely to realise that the namespace used by the deployer was different, the act of following that second instruction clobbers the ClusterRoleBinding (run-nfs-client-provisioner) setup by the Deployer.

Then subsequent runs of the Deployer that setup any PVs fail and leave PVCs in 'Pending' state, leading to Deployment failures.

Alan111S commented 1 month ago

Simple change in this file automation-roles/40-configure-infra/nfs-storage-class/vars/main.yml

Change:-

nfs_storage_class_project: cp-nfs

to

nfs_storage_class_project: nfs-provisioner

fketelaars commented 1 month ago

Making the following changes: