IBM / cloudpak-gitops

Deployment of IBM Cloud Paks using ArgoCD / Red Hat GitOps operator.
Apache License 2.0
22 stars 23 forks source link

feat: Reduce priority for "ibm-operator-catalog" catalog source #224

Closed nastacio closed 1 year ago

nastacio commented 1 year ago

Is your feature request related to a problem? Please describe. The ibm-operator-catalog catalog source added to the cluster by this repository has no priority set, which defaults to 0. With the documentation for all Cloud Paks indicating the possibility of the user adding "pinned" versions of catalog sources and not setting a priority either, that means the priority is being decided lexicographically (alphabetically.)

Describe the solution you'd like Parameterize the priority of the ibm-operator-catalog catalog source and default it to -1, so that eventual pinned catalog sources from other Cloud Paks take precedence over it.

Describe alternatives you've considered None.

Additional context

Official documentation on catalog source priorities and resolution: https://olm.operatorframework.io/docs/concepts/olm-architecture/dependency-resolution/#catalog-priority

Description of the .spec.priority field in CatalogSource:

Priority field assigns a weight to the catalog source to prioritize them so that it can be consumed by the dependency resolver. Usage: Higher weight indicates that this catalog source is preferred over lower weighted catalog sources during dependency resolution. The range of the priority value can go from positive to negative in the range of int32. The default value to a catalog source with unassigned priority would be 0. The catalog source with the same priority values will be ranked lexicographically based on its name.

nastacio commented 1 year ago

:tada: This issue has been resolved in version 0.20.0 :tada:

The release is available on GitHub release

Your semantic-release bot :package::rocket: