ensure that Cluster API keeps retrying to delete Nodes

Thomas Morinrequested to merge
kyverno-background-rbac-allow-machines into main

Workaround for #1431 (closed)

This MR integrates sylva-projects/sylva-elements/helm-charts/sylva-capi-cluster!393 (merged), which:

  • introduce parametrization of nodeDeletionTimeout to allow setting it to 0 (infinite retries)

The additions in sylva-core do the following:

  • a Kyverno policy to patch existing Machines
  • allow Kyverno background controller to touch CAPI Machines

co-authored with @feleouet, who analyze the issue with me and helped on the Kyverno bits

Edited by Thomas Morin

Merge request reports