inlined compute deprovision procedure in Removing a Hypervisor section

doc/source/operations/bifrost-hardware-inventory-management.rst

@@ -105,51 +105,9 @@ To deprovision an existing hypervisor, run:
 Removing a Hypervisor
 ---------------------
 
-To remove a hypervisor without replacing it, proceed as follows. For compute
-nodes, see :doc:`deprovision-compute` for a more detailed procedure.
+To remove a hypervisor without replacing it, proceed as follows.
 
-#. :ref:`Disable the hypervisor to avoid scheduling any new instances on it <taking-a-hypervisor-out-of-service>`
-#. :ref:`Evacuate all instances <evacuating-all-instances>`
-#. (optionally) Deprovision the hypervisor
-
-   .. code-block:: console
-
-      kayobe overcloud deprovision --limit <Hypervisor hostname>
-
-   .. warning::
-
-      Always use ``--limit`` with ``kayobe overcloud deprovision`` on a production
-      system. Running this command without a limit will deprovision all overcloud
-      hosts.
-
-#. Physically remove the node from the deployment
-
-#. Delete the node in Bifrost:
-
-   .. code-block:: console
-
-      docker exec -it bifrost_deploy bash
-      (bifrost-deploy)[root@seed bifrost-base]# export OS_CLOUD=bifrost
-      (bifrost-deploy)[root@seed bifrost-base]# openstack baremetal node delete <Hostname>
-
-#. Delete the compute service in OpenStack:
-
-   .. code-block:: console
-
-      openstack compute service list | grep <Hypervisor hostname>
-      openstack compute service delete <Service ID>
-
-#. Delete the network agents in OpenStack:
-
-   .. code-block:: console
-
-      openstack network agent list | grep <Hypervisor hostname>
-      openstack network agent delete <Agent IDs>
-
-#. Remove the node from the Kayobe configuration. Ensure the node is removed
-   from the inventory and ``network-allocation.yml``. Other configuration files
-   may also be removed, but this is dependent on the deployment. Recursive
-   ``grep`` can help here.
+.. include:: deprovision-compute.rst
 
 .. _evacuating-all-instances:
 

doc/source/operations/deprovision-compute.rst

@@ -1,21 +1,21 @@
-=============================
+#############################
 Deprovisioning a Compute Node
-=============================
+#############################
 
 This procedure removes a compute node from service, drains it, removes all
 traces of it from the control plane and Kayobe configuration, and finally
 deprovisions the hardware in Bifrost.
 
 Check the Hypervisor
-====================
+~~~~~~~~~~~~~~~~~~~~
 
 .. code-block:: console
 
    openstack compute service list --host <node>
    openstack server list --all-projects --host <node>
 
 Put It Into Maintenance Mode
-============================
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
 .. code-block:: console
 
@@ -29,7 +29,7 @@ Or use the OpenStack CLI:
    openstack compute service set --disable --disable-reason "decommission <ticket>" <node> nova-compute
 
 Drain the Node
-==============
+~~~~~~~~~~~~~~
 
 .. code-block:: console
 
@@ -52,7 +52,7 @@ Unpause any paused instances.
    openstack server migrate <id> --wait
 
 Confirm It's Empty
-==================
+~~~~~~~~~~~~~~~~~~
 
 On the to-be-decommissioned node:
 
@@ -61,15 +61,15 @@ On the to-be-decommissioned node:
    docker exec nova_libvirt virsh list --all
 
 Destroy Containers and Systemd Units
-====================================
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
 .. code-block:: console
 
    sudo systemctl disable --now $(systemctl list-units 'kolla*' --all --plain --no-legend | awk '{print $1}')
    docker stop $(docker ps -aq); docker rm $(docker ps -aq)
 
 Remove It From the Control Plane
-================================
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
 .. warning::
 
@@ -103,7 +103,7 @@ On a controller, check if the chassis exists and delete it:
    sudo docker exec ovn_sb_db ovn-sbctl --no-leader-only chassis-del <node>
 
 Remove It From Kayobe and Monitoring
-====================================
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
 Until this runs, Prometheus keeps scraping its exporters, and haproxy/fluentd
 keep referencing it.
@@ -119,7 +119,7 @@ Remove it from ``etc/kayobe/inventory/overcloud`` and any group vars, then:
    kayobe overcloud service reconfigure --kolla-limit controllers,monitoring
 
 Deprovision the Hardware
-========================
+~~~~~~~~~~~~~~~~~~~~~~~~
 
 From the seed.
 
@@ -132,7 +132,7 @@ From the seed.
    sudo docker exec -it bifrost_deploy bash -c 'export OS_CLOUD=bifrost; openstack baremetal node delete <node>'
 
 Verification
-============
+~~~~~~~~~~~~
 
 .. code-block:: console