Chapter 4. Adding physical machines as bare-metal nodes

Procedure

1. Create an inventory file to define the details of each node, for example, ironic-nodes.yaml.

2. For each node, define the node name and the address and credentials for the bare-metal driver. For details on the available properties for your enabled driver, see Bare metal drivers.

   nodes:
     - name: <node>
       driver: <driver>
       driver_info:
         <driver>_address: <ip>
         <driver>_username: <user>
         <driver>_password: <password>
         [<property>: <value>]

   • Replace <node> with the name of the node.
   • Replace <driver> with a supported bare-metal driver, for example,redfish.
   • Replace <ip> with the IP address of the Bare Metal controller.
   • Replace <user> with your username.
   • Replace <password> with your password.
   • Optional: Replace <property> with a driver property that you want to configure, and replace <value> with the value of the property. For information on the available properties, see Bare metal drivers.

3. Define the node properties and ports:

   nodes:
     - name: <node>
       ...
       properties:
         cpus: <cpu_count>
         cpu_arch: <cpu_arch>
         memory_mb: <memory>
         local_gb: <root_disk>
         root_device:
           serial: <serial>
         network_interface: <interface_type>
       ports:
         - address: <mac_address>

   • Replace <cpu_count> with the number of CPUs.
   • Replace <cpu_arch> with the type of architecture of the CPUs.
   • Replace <memory> with the amount of memory in MiB.
   • Replace <root_disk> with the size of the root disk in GiB. Only required when the machine has multiple disks.
   • Replace <serial> with the serial number of the disk that you want to use for deployment.

   • Optional: Include the network_interface property if you want to override the default network type of flat. You can change the network type to one of the following valid values:

     • neutron: Use to provide tenant-defined networking through the Networking service, where tenant networks are separated from each other and from the provisioning and cleaning provider networks. Required to create a provisioning network with IPv6.
     • noop: Use for standalone deployments where network switching is not required.
   • Replace <mac_address> with the MAC address of the NIC used to PXE boot.

4. Access the remote shell for the OpenStackClient pod from your workstation:

   $ oc rsh -n openstack openstackclient

5. Import the inventory file into the Bare Metal Provisioning service:

   $ openstack baremetal create ironic-nodes.yaml

   The nodes are now in the enroll state.

6. Wait for the extra network interface port configuration data to populate the Networking service (neutron). This process takes at least 60 seconds.

7. Set the provisioning state of each node to available:

   $ openstack baremetal node manage <node>
   $ openstack baremetal node provide <node>

   The Bare Metal Provisioning service cleans the node if you enabled node cleaning.

8. Check that the nodes are enrolled:

   $ openstack baremetal node list

   There might be a delay between enrolling a node and its state being shown.

9. Exit the openstackclient pod:

   $ exit

Procedure

1. Access the remote shell for the OpenStackClient pod from your workstation:

   $ oc rsh -n openstack openstackclient

2. Add a new node:

   $ openstack baremetal node create --driver <driver_name> --name <node_name>

   • Replace <driver_name> with the name of the driver, for example, redfish.
   • Replace <node_name> with the name of your new bare-metal node.
3. Note the UUID assigned to the node when it is created.

4. Update the node properties to match the hardware specifications on the node:

   $ openstack baremetal node set <node> \
     --property cpus=<cpu> \
     --property memory_mb=<ram> \
     --property local_gb=<disk> \
     --property cpu_arch=<arch>

   • Replace <node> with the ID of the bare metal node.
   • Replace <cpu> with the number of CPUs.
   • Replace <ram> with the RAM in MB.
   • Replace <disk> with the disk size in GB.
   • Replace <arch> with the architecture type.

5. Optional: Set the network_interface property to override the default network type of flat:

   $ openstack baremetal node set <node> --network-interace <network_interface>

   • Replace <network_interface> with one of the following valid network types:

     • neutron: Use to provide tenant-defined networking through the Networking service, where tenant networks are separated from each other and from the provisioning and cleaning provider networks. Required to create a provisioning network with IPv6.
     • noop: Use for standalone deployments where network switching is not required.

6. Optional: If you have multiple disks, set the root device hints to inform the deploy ramdisk which disk to use for deployment:

   $ openstack baremetal node set <node> \
     --property root_device='{"<property>": "<value>"}'

   • Replace <node> with the ID of the bare metal node.

   • Replace <property> and <value> with details about the disk that you want to use for deployment, for example root_device='{"size": "128"}'

     RHOSP supports the following properties:

     • model (String): Device identifier.
     • vendor (String): Device vendor.
     • serial (String): Disk serial number.
     • hctl (String): Host:Channel:Target:Lun for SCSI.
     • size (Integer): Size of the device in GB.
     • wwn (String): Unique storage identifier.
     • wwn_with_extension (String): Unique storage identifier with the vendor extension appended.
     • wwn_vendor_extension (String): Unique vendor storage identifier.
     • rotational (Boolean): True for a rotational device (HDD), otherwise false (SSD).

     • name (String): The name of the device, for example: /dev/sdb1 Use this property only for devices with persistent names.

       Note

       If you specify more than one property, the device must match all of those properties.

7. Inform the Bare Metal Provisioning service of the node network card by creating a port with the MAC address of the NIC on the provisioning network:

   $ openstack baremetal port create --node <node_uuid> <mac_address>

   • Replace <node> with the unique ID of the bare metal node.
   • Replace <mac_address> with the MAC address of the NIC used to PXE boot.

8. Validate the configuration of the node:

   $ openstack baremetal node validate <node>
   +------------+--------+---------------------------------------------+
   | Interface  | Result | Reason                                      |
   +------------+--------+---------------------------------------------+
   | boot       | False  | Cannot validate image information for node  |
   |            |        | a02178db-1550-4244-a2b7-d7035c743a9b        |
   |            |        | because one or more parameters are missing  |
   |            |        | from its instance_info. Missing are:        |
   |            |        | ['ramdisk', 'kernel', 'image_source']       |
   | console    | None   | not supported                               |
   | deploy     | False  | Cannot validate image information for node  |
   |            |        | a02178db-1550-4244-a2b7-d7035c743a9b        |
   |            |        | because one or more parameters are missing  |
   |            |        | from its instance_info. Missing are:        |
   |            |        | ['ramdisk', 'kernel', 'image_source']       |
   | inspect    | None   | not supported                               |
   | management | True   |                                             |
   | network    | True   |                                             |
   | power      | True   |                                             |
   | raid       | True   |                                             |
   | storage    | True   |                                             |
   +------------+--------+---------------------------------------------+

   The validation output Result indicates the following:

   • False: The interface has failed validation. If the reason provided includes missing the instance_info parameters [\'ramdisk', \'kernel', and \'image_source'], this might be because the Compute service populates those missing parameters at the beginning of the deployment process, therefore they have not been set at this point. If you are using a whole disk image, then you might need to only set image_source to pass the validation.
   • True: The interface has passed validation.
   • None: The interface is not supported for your driver.

9. Exit the openstackclient pod:

   $ exit

Procedure

1. Access the remote shell for the OpenStackClient pod from your workstation:

   $ oc rsh -n openstack openstackclient

2. Set the Bare Metal service boot interface to redfish-virtual-media:

   $ openstack baremetal node set --boot-interface redfish-virtual-media <node_name>

   • Replace <node_name> with the name of the node.

3. For UEFI nodes, define the EFI System Partition (ESP) image:

   $ openstack baremetal node set --driver-info bootloader=<esp_image> <node>

   • Replace <esp_image> with the image UUID or URL for the ESP image.
   • Replace <node> with the name of the node.
   Note

   For BIOS nodes, do not complete this step.

4. Create a port on the bare-metal node and associate the port with the MAC address of the NIC on the bare metal node:

   $ openstack baremetal port create --pxe-enabled True --node <node_uuid> <mac_address>

   • Replace <node_uuid> with the UUID of the bare-metal node.
   • Replace <mac_address> with the MAC address of the NIC on the bare-metal node.

5. Exit the openstackclient pod:

   $ exit

Procedure

1. Access the remote shell for the OpenStackClient pod from your workstation:

   $ oc rsh -n openstack openstackclient

2. Retrieve a list of your nodes to identify their UUIDs:

   $ openstack baremetal node list

3. Tag each bare-metal node with a custom bare-metal resource class:

   $ openstack baremetal node set \
    --resource-class baremetal.<CUSTOM> <node>

   • Replace <CUSTOM> with a string that identifies the purpose of the resource class. For example, set to GPU to create a custom GPU resource class that you can use to tag bare metal nodes that you want to designate for GPU workloads.
   • Replace <node> with the ID of the bare metal node.

4. Create a flavor for bare-metal instances:

   $ openstack flavor create --id auto \
    --ram <ram_size_mb> --disk <disk_size_gb> \
    --vcpus <no_vcpus> baremetal

   • Replace <ram_size_mb> with the RAM of the bare metal node, in MB.
   • Replace <disk_size_gb> with the size of the disk on the bare metal node, in GB.

   • Replace <no_vcpus> with the number of CPUs on the bare metal node.

     Note

     These properties are not used for scheduling instances. However, the Compute scheduler does use the disk size to determine the root partition size.

5. Associate the flavor for bare-metal instances with the custom resource class:

   $ openstack flavor set \
    --property resources:CUSTOM_BAREMETAL_<CUSTOM>=1 \
    baremetal

   To determine the name of a custom resource class that corresponds to a resource class of a bare-metal node, convert the resource class to uppercase, replace each punctuation mark with an underscore, and prefix with CUSTOM_.

   Note

   A flavor can request only one instance of a bare-metal resource class.

6. Set the following flavor properties to prevent the Compute scheduler from using the bare-metal flavor properties to schedule instances:

   $ openstack flavor set \
    --property resources:VCPU=0 \
    --property resources:MEMORY_MB=0 \
    --property resources:DISK_GB=0 baremetal

7. Verify that the new flavor has the correct values:

   $ openstack flavor list

8. Exit the openstackclient pod:

   $ exit