Is your feature request related to a problem? Please describe.
The current cluster registration flow in pull mode requires multiple manual steps:
- register → generate manifest → manually apply it to the target cluster.
This is time-consuming and requires context switching, especially when working with multiple clusters. It also feels inconsistent compared to push mode, which already simplifies parts of this workflow.
Describe the solution you'd like
Introduce an option like --auto-apply for pull mode that automatically applies the generated manifest to the target (managed) cluster. If the user already has access and context to the managed cluster.
Example:
sveltosctl register cluster \
--namespace=cluster-0 \
--cluster=cluster-0 \
--fleet-cluster-context=cluster-0 \
--labels=kro=enabled \
--pullmode \
--auto-apply
Additionally, since the user is already executing the sveltosctl command, the CLI could handle two contexts:
- One for the management cluster (registration step)
- One for the managed cluster (applying the manifest)
Proposed additional flow:
- Register cluster against the management cluster
- Generate the manifest
- Automatically apply it to the managed cluster
This would remove the need for manual context switching and applying manifests.
Describe alternatives you've considered
- Keeping the current workflow (manual apply after manifest generation)
- Printing the manifest and applying it via separate tooling/scripts
- Using push mode where applicable, though it does not fully replace pull mode use cases
Additional context
-
It would be useful to support multiple modes:
- Print manifest only
- Print + manual apply
- Print + auto-apply
-
Aligning pull mode behavior more closely with push mode would improve UX consistency
-
This change would significantly speed up cluster onboarding and reduce operational friction
Is your feature request related to a problem? Please describe.
The current cluster registration flow in pull mode requires multiple manual steps:
This is time-consuming and requires context switching, especially when working with multiple clusters. It also feels inconsistent compared to push mode, which already simplifies parts of this workflow.
Describe the solution you'd like
Introduce an option like
--auto-applyfor pull mode that automatically applies the generated manifest to the target (managed) cluster. If the user already has access and context to the managed cluster.Example:
Additionally, since the user is already executing the
sveltosctlcommand, the CLI could handle two contexts:Proposed additional flow:
This would remove the need for manual context switching and applying manifests.
Describe alternatives you've considered
Additional context
It would be useful to support multiple modes:
Aligning pull mode behavior more closely with push mode would improve UX consistency
This change would significantly speed up cluster onboarding and reduce operational friction