You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Adding option to selectively mirror operators in mirror_ocp role can help reducing mirror file size.
For example, we don't need NVIDIA GPU Operator for Manage only MAS instance.
Also, adding LVM Storage Operator would be nice for single node configuration.
The text was updated successfully, but these errors were encountered:
Although I like the idea of it being configurable, I don't think we are likely to ever support this, we are already stepping beyond what we originally wanted to own by providing a pre-defined definition of a MAS subset of the Red Hat operator catalogs. For cases where a customer wants to further customize I think that's going to lie within their own processes & procedures for managing their cluster; we want to keep mirror_ocp and mirror-redhat-images relatively straight forward as it's more a bonus than a core part of what we're wanting to focus on (which is MAS). It exists because many customers have struggled with the base tooling provided by Red Hat, but to be able to maintain it we need to keep it relatively simple with minimal variations.
We can certainly look at adding the LVM Storage Operator to the set though.
Adding option to selectively mirror operators in
mirror_ocp
role can help reducing mirror file size.For example, we don't need
NVIDIA GPU Operator
for Manage only MAS instance.Also, adding
LVM Storage Operator
would be nice for single node configuration.The text was updated successfully, but these errors were encountered: