Added a PVC to the spire-server deployment in order to persist the spire server data store (especially node and workload registrations) across pod restarts. This would fail if the target cluster does not have a default storage class, and we could handle that with some logic in the deploy-spire-and-csi-driver.sh script, but in the majority of cases, storage will be available, and certainly the spire-server needs persistent storage in any real-world use case. So I think it's fair to expect a PVC creation to succeed, for this example.
Added a PVC to the spire-server deployment in order to persist the spire server data store (especially node and workload registrations) across pod restarts. This would fail if the target cluster does not have a default storage class, and we could handle that with some logic in the deploy-spire-and-csi-driver.sh script, but in the majority of cases, storage will be available, and certainly the spire-server needs persistent storage in any real-world use case. So I think it's fair to expect a PVC creation to succeed, for this example.
Signed-off-by: Abe Sharp abe@hpe.com