Using latest branch (at the time 1.16.3), in point 3e, it indicates to use kubectl to apply config/manager/manager.yaml to the Kubernetes cluster. This file, however, includes a Deployment definition that does not include a namespace name in the metadata:
Unless the user has specified a default Namespace for kubectl of ibm-common-services, this will be deployed to whatever their default Namespace happens to be (typically "default", but could be other values as well).
The Deployment, however, references a serviceAccountName: ibm-licensing-operator , which is defined in config/rbac/service_account.yaml - and this ServiceAccount does specify the ibm-common-services namespace. As a result, the Deployment does not work.
To be consistent with the other manifests, the Deployment defined in config/manager/manager.yaml should have a namespace attribute of ibm-common-services added to the metadata.
The instructions here: https://www.ibm.com/docs/en/cpfs?topic=software-offline-installation
Using latest branch (at the time 1.16.3), in point 3e, it indicates to use kubectl to apply config/manager/manager.yaml to the Kubernetes cluster. This file, however, includes a Deployment definition that does not include a namespace name in the metadata:
Unless the user has specified a default Namespace for kubectl of ibm-common-services, this will be deployed to whatever their default Namespace happens to be (typically "default", but could be other values as well).
The Deployment, however, references a
serviceAccountName: ibm-licensing-operator
, which is defined in config/rbac/service_account.yaml - and this ServiceAccount does specify the ibm-common-services namespace. As a result, the Deployment does not work.To be consistent with the other manifests, the Deployment defined in config/manager/manager.yaml should have a namespace attribute of ibm-common-services added to the metadata.