Recovering from expired control plane certificates

Follow this procedure to recover from a situation where your control plane certificates have expired.

  • SSH access to master hosts.

  1. Access a master host with an expired certificate as the root user.

  2. Obtain the cluster-kube-apiserver-operator image reference for a release.

    # RELEASE_IMAGE=<release_image> (1)
    1 An example value for <release_image> is quay.io/openshift-release-dev/ocp-release:4.1.0.
    # KAO_IMAGE=$( oc adm release info --registry-config='/var/lib/kubelet/config.json' "${RELEASE_IMAGE}" --image-for=cluster-kube-apiserver-operator )
  3. Pull the cluster-kube-apiserver-operator image.

    # podman pull --authfile=/var/lib/kubelet/config.json "${KAO_IMAGE}"
  4. Create a recovery API server.

    # podman run -it --network=host -v /etc/kubernetes/:/etc/kubernetes/:Z --entrypoint=/usr/bin/cluster-kube-apiserver-operator "${KAO_IMAGE}" recovery-apiserver create
  5. Run the export KUBECONFIG command from the output of the above command, which is needed for the oc commands later in this procedure.

    # export KUBECONFIG=/<path_to_recovery_kubeconfig>/admin.kubeconfig
  6. Wait for the recovery API server to come up.

    # until oc get namespace kube-system 2>/dev/null 1>&2; do echo 'Waiting for recovery apiserver to come up.'; sleep 1; done
  7. Run the regenerate-certificates command. It fixes the certificates in the API, overwrites the old certificates on the local drive, and restarts static Pods to pick them up.

    # podman run -it --network=host -v /etc/kubernetes/:/etc/kubernetes/:Z --entrypoint=/usr/bin/cluster-kube-apiserver-operator "${KAO_IMAGE}" regenerate-certificates
  8. After the certificates are fixed in the API, use the following commands to force new rollouts for the control plane. It will reinstall itself on the other nodes because the kubelet is connected to API servers using an internal load balancer.

    # oc patch kubeapiserver cluster -p='{"spec": {"forceRedeploymentReason": "recovery-'"$( date --rfc-3339=ns )"'"}}' --type=merge
    # oc patch kubecontrollermanager cluster -p='{"spec": {"forceRedeploymentReason": "recovery-'"$( date --rfc-3339=ns )"'"}}' --type=merge
    # oc patch kubescheduler cluster -p='{"spec": {"forceRedeploymentReason": "recovery-'"$( date --rfc-3339=ns )"'"}}' --type=merge
  9. Create a bootstrap kubeconfig with a valid user.

    1. Run the recover-kubeconfig.sh script and save the output to a file called kubeconfig.

      # recover-kubeconfig.sh > kubeconfig
    2. Copy the kubeconfig file to all master hosts and move it to /etc/kubernetes/kubeconfig.

    3. Get the CA certificate used to validate connections from the API server.

      # oc get configmap kube-apiserver-to-kubelet-client-ca -n openshift-kube-apiserver-operator --template='{{ index .data "ca-bundle.crt" }}' > /etc/kubernetes/ca.crt
    4. Copy the /etc/kubernetes/ca.crt file to all other master hosts and nodes.

    5. Add the machine-config-daemon-force file to all master hosts and nodes to force the Machine Config Daemon to accept this certificate update.

      # touch /run/machine-config-daemon-force
  10. Recover the kubelet on all masters.

    1. On a master host, stop the kubelet.

      # systemctl stop kubelet
    2. Delete stale kubelet data.

      # rm -rf /var/lib/kubelet/pki /var/lib/kubelet/kubeconfig
    3. Restart the kubelet.

      # systemctl start kubelet
    4. Repeat these steps on all other master hosts.

  11. If necessary, recover the kubelet on the worker nodes.

    After the master nodes are restored, the worker nodes might restore themselves. You can verify this by running the oc get nodes command. If the worker nodes are not listed, then perform the following steps on each worker node.

    1. Stop the kubelet.

      # systemctl stop kubelet
    2. Delete stale kubelet data.

      # rm -rf /var/lib/kubelet/pki /var/lib/kubelet/kubeconfig
    3. Restart the kubelet.

      # systemctl start kubelet
  12. Approve the pending node-bootstrapper certificates signing requests (CSRs).

    1. Get the list of current CSRs.

      # oc get csr
    2. Review the details of a CSR to verify it is valid.

      # oc describe csr <csr_name> (1)
      1 <csr_name> is the name of a CSR from the list of current CSRs.
    3. Approve each valid CSR.

      # oc adm certificate approve <csr_name>

      Be sure to approve all pending node-bootstrapper CSRs.

  13. Destroy the recovery API server because it is no longer needed.

    # podman run -it --network=host -v /etc/kubernetes/:/etc/kubernetes/:Z --entrypoint=/usr/bin/cluster-kube-apiserver-operator "${KAO_IMAGE}" recovery-apiserver destroy

    Wait for the control plane to restart and pick up the new certificates. This might take up to 10 minutes.