[bug] K0rdent deployment is stuck: CAPI is not ready

[eromanova]

Describe the bug
Reproducibility: rare (faced only once).

Sometimes the CAPI installation happens before the CAPI operator is fully functional which leads to the following error:

error: "HelmRelease kcm-system/capi is not yet ready: Helm install failed
          for release kcm-system/capi with chart [email protected]: 1 error occurred:\n\t*
          Internal error occurred: failed calling webhook \"vcoreprovider.kb.io\":
          failed to call webhook: Post \"[https://capi-operator-webhook-service.kcm-system.svc:443/mutate-operator-cluster-x-k8s-io-v1alpha2-coreprovider?timeout=10s](https://capi-operator-webhook-service.kcm-system.svc/mutate-operator-cluster-x-k8s-io-v1alpha2-coreprovider?timeout=10s)\":
          dial tcp 10.96.119.28:443: i/o timeout"

The capi HelmRelease is in a Failed state and the installation is not retried. Thus, the installation of all other components is stuck.

Workaround

1. Remove the capi HelmRelease

To Reproduce
Steps to reproduce the behavior:

1. Install k0rdent v0.1.0 using helm install

Expected behavior
All K0rdent components are up and running

[prazumovsky]

Remove the capi HelmRelease

Any change to capi helmrelease spec or corresponding helmrepo spec is also working as retry btw