And use jq to extract the cluster ID from the metadata file, since that's where openshift-install destroy cluster gets the information.
Also, one small quibble is that most, not all, resources are tagged with openshiftClusterID. Those resources are also tagged with kubernetes.io/cluster/{clusterName}: owned, and a few resources are only tagged with the kubernetes.io tag (because they are created by upstream Kubernetes tooling that does not provide knobs to add additional tags, or because an OpenShift operator doesn't yet use existing knobs). And some resources aren't tagged at all, because AWS only supports tagging for some resource types. Still, openshiftClusterID gets the general idea across, so I've just removed "all" and kept the weed-wading to this message ;).
And use
jqto extract the cluster ID from the metadata file, since that's whereopenshift-install destroy clustergets the information.Also, one small quibble is that most, not all, resources are tagged with
openshiftClusterID. Those resources are also tagged withkubernetes.io/cluster/{clusterName}: owned, and a few resources are only tagged with thekubernetes.iotag (because they are created by upstream Kubernetes tooling that does not provide knobs to add additional tags, or because an OpenShift operator doesn't yet use existing knobs). And some resources aren't tagged at all, because AWS only supports tagging for some resource types. Still,openshiftClusterIDgets the general idea across, so I've just removed "all" and kept the weed-wading to this message ;).