Docs update needed: Deleting radius-system namespace may hang #1524
Description
What content needs to be created or modified?
The documentation should be updated to include troubleshooting steps and expected behavior when deleting the radius-system namespace. According to this issue in the radius repo, users may experience indefinite hangs due to resource deletion failures and remaining pods. Documentation should address:
- What happens during deletion
- Why the namespace may hang (e.g., remaining pods, API discovery failures)
- How to resolve or troubleshoot the problem
Describe the solution you'd like
A new troubleshooting section should be added for the radius-system namespace deletion scenario, including:
- Step-by-step guidance for users facing this problem
- How to check for remaining resources
- How to force delete stuck resources
- How to interpret error messages such as "Discovery failed for some groups" and "Failed to delete all resource types"
Difference between rad uninstall kubernetes --purge and without --purge
Add a section explaining:
- What happens when running
rad uninstall kubernetes --purgevs. without--purge - What data and resources are left behind when not running with
--purge(e.g., some CRDs, persistent data, configmaps, or other resources) - Best practices for users who want a complete clean-up
Where should the new material be placed?
Suggest placement in the Kubernetes installation and management docs, under troubleshooting or FAQ sections.
The associated pull request from the Radius code repo, if applicable
Reference issue: radius-project/radius#10255
Additional context
See the linked issue for details, example output, and observed behavior. This update will help users avoid confusion and successfully clean up resources when uninstalling Radius.