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
Initialize garden resources in the cluster via garden plugins local-kubernetes cluster-init
Upgrade to the 0.13.0 version of the CLI
Run garden validate - this triggers upgrades of garden managed resources including the nginx controller
Expected results: No errors and components are updated
Actual results:
Resource: "batch/v1, Resource=jobs", GroupVersionKind: "batch/v1, Kind=Job"Name: "ingress-nginx-admission-patch", Namespace: "ingress-nginx"...TopologySpreadConstraints:[]core.TopologySpreadConstraint(nil), OS:(*core.PodOS)(nil)}}: field is immutable...Failed resolving one or more providers:- local-kubernetes
The text was updated successfully, but these errors were encountered:
So my initial thought is that we have two options:
Wrap the kind manifests in a Helm chart and use a helm Deploy.
Stick to using a kubernetes Deploy, but detect the immutable field change error here, and force-deploy the ingress controller when that happens (notifying the user that we're about to do so with a warning message).
In the meantime, a workaround on kind would be to uninstall the system services using garden plugins kubernetes uninstall-garden-services on 0.12, and then running garden plugins local-kubernetes cluster-init on 0.13 to install the ingress controller from scratch (i.e. without attempting to patch the already-installed one).
Note
This is a copy/paste issue from a Discord thread
Garden Bonsai (0.13) Bug
After upgrading to the 0.13 version of the CLI, running any command tries to update the nginx ingress controller and fails.
Steps to replicate:
Expected results: No errors and components are updated
Actual results:
The text was updated successfully, but these errors were encountered: