Kyverno HelmRelease is failed on bootstrap cluster - slow platform
This is an issue that has been observed multiple times but occurs randomly on a baremetal platform. The Kyverno HelmRelease on the bootstrap cluster doesn't install correctly or is not recognized as installed.
2025/06/02 17:42:01.478425 Kustomization/kyverno state changed: HealthCheckFailed - health check failed after 33.623904ms: failed early due to stalled resources: [HelmRelease/sylva-system/kyverno status: 'Failed']
2025/06/02 17:42:31.516471 Kustomization/kyverno state changed: Progressing - Reconciliation in progress
2025/06/02 17:42:31.777425 Kustomization/kyverno state changed: HealthCheckFailed - health check failed after 40.846334ms: failed early due to stalled resources: [HelmRelease/sylva-system/kyverno status: 'Failed']
2025/06/02 17:42:49.279899 Command timeout exceeded
All Kyverno pods are running. It seems that the default timeout of the helmrelease is too short (5minutes). The mechanism for retrying the HelmRelease cannot work in this case, as it is based on Kyverno policies. (https://gitlab.com/sylva-projects/sylva-core/-/blob/main/kustomize-units/kyverno-policies/generic/components/force-reconcile-helmreleases/force-reconcile-helmreleases.yaml)
Edited by Remi Le Trocquer