Chapter 4. Configuring the overcloud

Use Red Hat OpenStack Platform (RHOSP) director to install and configure spine leaf networking in the RHOSP overcloud. The high-level steps are:

  1. Define the overcloud networks for each leaf.
  2. Create a composable role for each leaf and attach the composable network to each respective role.
  3. Create a unique NIC configuration for each role.
  4. Change bridge mappings so that each leaf routes traffic through the specific bridge or VLAN on that leaf.
  5. Define virtual IPs (VIPs) for your overcloud endpoints, and identify the subnet for each VIP.
  6. Provision your overcloud networks and overcloud VIPs.

  7. Register the bare metal nodes in your overcloud.

    Note

    Skip steps 7, 8, and 9 if you are using pre-provisioned bare metal nodes.

  8. Introspect the bare metal nodes in your overcloud.
  9. Provision bare metal nodes.
  10. Deploy your overcloud using the configuration you set in the earlier steps.

4.1. Defining the leaf networks

The Red Hat OpenStack Platform (RHOSP) director creates the overcloud leaf networks from a YAML-formatted, custom network definition file that you construct. This custom network definition file lists each composable network and its attributes and also defines the subnets needed for each leaf.

Complete the following steps to create a YAML-formatted, custom network definition file that contains the specifications for your spine-leaf network on the overcloud. Later, the provisioning process creates a heat environment file from your network definition file that you include when you deploy your RHOSP overcloud.

Prerequisites

  • Access to the undercloud host and credentials for the stack user.

Procedure

  1. Log in to the undercloud host as the stack user.

  2. Source the stackrc undercloud credentials file: 

    $ source ~/stackrc

  3. Create a templates directory under /home/stack: 

    $ mkdir /home/stack/templates

  4. Copy the default network definition template, routed-networks.yaml, to your custom templates directory:

    Example

    $ cp /usr/share/openstack-tripleo-heat-templates/network-data-samples/\
routed-networks.yaml \
/home/stack/templates/spine-leaf-networks-data.yaml

  5. Edit your copy of the network definition template to define each base network and each of the associated leaf subnets as a composable network item.

    Tip

    For information, see Network definition file configuration options in the Installing and managing Red Hat OpenStack Platform with director guide.

    Example

    The following example demonstrates how to define the Internal API network and its leaf networks: 

    - name: InternalApi
  name_lower: internal_api
  vip: true
  mtu: 1500
  subnets:
    internal_api_subnet:
      ip_subnet: 172.16.32.0/24
      gateway_ip: 172.16.32.1
      allocation_pools:
        - start: 172.16.32.4
          end: 172.16.32.250
      vlan: 20
    internal_api_leaf1_subnet:
      ip_subnet: 172.16.33.0/24
      gateway_ip: 172.16.33.1
      allocation_pools:
        - start: 172.16.33.4
          end: 172.16.33.250
      vlan: 30
    internal_api_leaf2_subnet:
      ip_subnet: 172.16.34.0/24
      gateway_ip: 172.16.34.1
      allocation_pools:
        - start: 172.16.34.4
          end: 172.16.34.250
      vlan: 40
Note

You do not define the Control Plane networks in your custom network definition template, because the undercloud has already created these networks. However, you must set the parameters manually so that the overcloud can configure the NICs accordingly. For more information, see Configuring routed spine-leaf in the undercloud.

Note

RHOSP does not perform automatic validation of the network subnet and allocation_pools values. Ensure that you define these values consistently and that they do not conflict with existing networks.

Note

Add the vip parameter and set the value to true for the networks that host the Controller-based services. In this example, the InternalApi network contains these services.

Next steps

  1. Note the path and file name of the custom network definition file that you have created. You need this information later when you provision your networks for the RHOSP overcloud.
  2. Proceed to the next step Defining leaf roles and attaching networks.

Additional resources

4.2. Defining leaf roles and attaching networks

The Red Hat OpenStack Platform (RHOSP) director creates a composable role for each leaf and attaches the composable network to each respective role from a roles template that you construct. Start by copying the default Controller, Compute, and Ceph Storage roles from the director core templates, and modifying these to meet your environment’s needs. After you have created all of the individual roles, you run the openstack overcloud roles generate command to concatenate them into one large custom roles data file.

Prerequisites

  • Access to the undercloud host and credentials for the stack user.

Procedure

  1. Log in to the undercloud host as the stack user.

  2. Source the stackrc undercloud credentials file: 

    $ source ~/stackrc

  3. Copy the default roles for Controller, Compute, and Ceph Storage roles that ship with RHOSP to the home directory of the stack user. Rename the files to indicate that they are leaf 0: 

    $ cp /usr/share/openstack-tripleo-heat-templates/roles/Controller.yaml ~/roles/Controller0.yaml
$ cp /usr/share/openstack-tripleo-heat-templates/roles/Compute.yaml ~/roles/Compute0.yaml
$ cp /usr/share/openstack-tripleo-heat-templates/roles/CephStorage.yaml ~/roles/CephStorage0.yaml

  4. Copy the leaf 0 files to create your leaf 1 and leaf 2 files: 

    $ cp ~/roles/Compute0.yaml ~/roles/Compute1.yaml
$ cp ~/roles/Compute0.yaml ~/roles/Compute2.yaml
$ cp ~/roles/CephStorage0.yaml ~/roles/CephStorage1.yaml
$ cp ~/roles/CephStorage0.yaml ~/roles/CephStorage2.yaml

  5. Edit the parameters in each file to align with their respective leaf parameters.

    Tip

    For information about the various parameters in a roles data template, see Examining role parameters in the Installing and managing Red Hat OpenStack Platform with director guide.

    Example - ComputeLeaf0

    - name: ComputeLeaf0
  HostnameFormatDefault: '%stackname%-compute-leaf0-%index%'

    Example - CephStorageLeaf0

    - name: CephStorageLeaf0
  HostnameFormatDefault: '%stackname%-cephstorage-leaf0-%index%'

  6. Edit the network parameter in the leaf 1 and leaf 2 files so that they align with the respective leaf network parameters.

    Example - ComputeLeaf1

    - name: ComputeLeaf1
  networks:
    InternalApi:
      subnet: internal_api_leaf1
    Tenant:
      subnet: tenant_leaf1
    Storage:
      subnet: storage_leaf1

    Example - CephStorageLeaf1

    - name: CephStorageLeaf1
  networks:
    Storage:
      subnet: storage_leaf1
    StorageMgmt:
      subnet: storage_mgmt_leaf1

    Note

    This applies only to leaf 1 and leaf 2. The network parameter for leaf 0 retains the base subnet values, which are the lowercase names of each subnet combined with a _subnet suffix. For example, the Internal API for leaf 0 is internal_api_subnet.

  7. When your role configuration is complete, run the overcloud roles generate command to generate the full roles data file.

    Example

    $ openstack overcloud roles generate --roles-path ~/roles -o spine-leaf-roles-data.yaml Controller Compute Compute1 Compute2 CephStorage CephStorage1 CephStorage2

    This creates one custom roles data file that includes all of the custom roles for each respective leaf network.

Next steps

  1. Note the path and file name of the custom roles data file created by the overcloud roles generate command. You use this path later when you deploy your overcloud.
  2. Proceed to the next step Creating a custom NIC configuration for leaf roles.

Additional resources

4.3. Creating a custom NIC configuration for leaf roles

Each role that the Red Hat OpenStack Platform (RHOSP) director creates requires a unique NIC configuration. Complete the following steps to create a custom set of NIC templates and a custom environment file that maps the custom templates to the respective role.

Prerequisites

  • Access to the undercloud host and credentials for the stack user.
  • You have a custom network definition file.
  • You have a custom roles data file.

Procedure

  1. Log in to the undercloud host as the stack user.

  2. Source the stackrc undercloud credentials file: 

    $ source ~/stackrc

  3. Copy the content from one of the default NIC templates to create a custom template for your NIC configuration.

    Example

    In this example, the single-nic-vlans NIC template is copied to use for a custom template for your NIC configuration: 

    $ cp -r /usr/share/ansible/roles/tripleo_network_config/\
templates/single-nic-vlans/* /home/stack/templates/spine-leaf-nics/.

  4. In each of the NIC templates that you created in the earlier step, change the NIC configuration to match the specifics for your spine-leaf topology.

    Example

    {% set mtu_list = [ctlplane_mtu] %}
{% for network in role_networks %}
{{ mtu_list.append(lookup('vars', networks_lower[network] ~ '_mtu')) }}
{%- endfor %}
{% set min_viable_mtu = mtu_list | max %}
network_config:
- type: ovs_bridge
  name: {{ neutron_physical_bridge_name }}
  mtu: {{ min_viable_mtu }}
  use_dhcp: false
  dns_servers: {{ ctlplane_dns_nameservers }}
  domain: {{ dns_search_domains }}
  addresses:
  - ip_netmask: {{ ctlplane_ip }}/{{ ctlplane_subnet_cidr }}
  routes: {{ ctlplane_host_routes }}
  members:
  - type: interface
    name: nic1
    mtu: {{ min_viable_mtu }}
    # force the MAC address of the bridge to this interface
    primary: true
{% for network in role_networks %}
  - type: vlan
    mtu: {{ lookup('vars', networks_lower[network] ~ '_mtu') }}
    vlan_id: {{ lookup('vars', networks_lower[network] ~ '_vlan_id') }}
    addresses:
    - ip_netmask:
        {{ lookup('vars', networks_lower[network] ~ '_ip') }}/{{ lookup('vars', networks_lower[network] ~ '_cidr') }}
    routes: {{ lookup('vars', networks_lower[network] ~ '_host_routes') }}
{% endfor %}

    Tip

    For more information, see Custom network interface templates in the Installing and managing Red Hat OpenStack Platform with director guide.

  5. Create a custom environment file, such as spine-leaf-nic-roles-map.yaml, that contains a parameter_defaults section that maps the custom NIC templates to each custom role. 

    parameter_defaults:
  %%ROLE%%NetworkConfigTemplate: <path_to_ansible_jinja2_nic_config_file>

    Example

    parameter_defaults:
  Controller0NetworkConfigTemplate: '/home/stack/templates/spine-leaf-nics/single-nic-vlans.j2'
  Controller1NetworkConfigTemplate: '/home/stack/templates/spine-leaf-nics/single-nic-vlans.j2'
  Controller2NetworkConfigTemplate: '/home/stack/templates/spine-leaf-nics/single-nic-vlans.j2'
  ComputeLeaf0NetworkConfigTemplate: '/home/stack/templates/spine-leaf-nics/single-nic-vlans.j2'
  ComputeLeaf1NetworkConfigTemplate: '/home/stack/templates/spine-leaf-nics/single-nic-vlans.j2'
  ComputeLeaf2NetworkConfigTemplate: '/home/stack/templates/spine-leaf-nics/single-nic-vlans.j2'
  CephStorage0NetworkConfigTemplate: '/home/stack/templates/spine-leaf-nics/single-nic-vlans.j2'
  CephStorage1NetworkConfigTemplate: '/home/stack/templates/spine-leaf-nics/single-nic-vlans.j2'
  CephStorage2NetworkConfigTemplate: '/home/stack/templates/spine-leaf-nics/single-nic-vlans.j2'

Next steps

  1. Note the path and file name of your custom NIC templates and the custom environment file that maps the custom NIC templates to each custom role. You use this path later when you deploy your overcloud.
  2. Proceed to the next step Configuring the leaf networks.

Additional resources

4.4. Configuring the leaf networks

In a spine leaf architecture, each leaf routes traffic through the specific bridge or VLAN on that leaf, which is often the case with edge computing scenarios. So, you must change the default mappings where the Red Hat OpenStack Platform (RHOSP) Controller and Compute network configurations use a br-ex bridge.

The RHOSP director creates the control plane network during undercloud creation. However, the overcloud requires access to the control plane for each leaf. To enable this access, you must define additional parameters in your deployment.

Complete the following steps to create a custom network environment file that contains the separate network mappings and sets access to the control plane networks for the overcloud.

Prerequisites

  • Access to the undercloud host and credentials for the stack user.

Procedure

  1. Log in to the undercloud host as the stack user.

  2. Source the stackrc undercloud credentials file: 

    $ source ~/stackrc

  3. In a new custom environment file, such as spine-leaf-ctlplane.yaml, create a parameter_defaults section and set the NeutronBridgeMappings parameter for each leaf that uses the default br-ex bridge.

    Important

    The name of the custom environment file that you create to contain your network definition must end in either .yaml or .template.

    • For flat network mappings, list each leaf in the NeutronFlatNetworks parameter and set the NeutronBridgeMappings parameter for each leaf:

      Example

      parameter_defaults:
  NeutronFlatNetworks: leaf0,leaf1,leaf2
  Controller0Parameters:
    NeutronBridgeMappings: "leaf0:br-ex"

  Controller1Parameters:
    NeutronBridgeMappings: "leaf0:br-ex"

  Controller2Parameters:
    NeutronBridgeMappings: "leaf0:br-ex"

  Compute0Parameters:
    NeutronBridgeMappings: "leaf0:br-ex"

  Compute1Parameters:
    NeutronBridgeMappings: "leaf1:br-ex"

  Compute2Parameters:
    NeutronBridgeMappings: "leaf2:br-ex"

      Tip

      For more information, see Chapter 17. Networking (neutron) Parameters in the Overcloud parameters guide

    • For VLAN network mappings, add vlan to NeutronNetworkType, and by using NeutronNetworkVLANRanges, map VLANs for the leaf networks:

      Example

      parameter_defaults:
  NeutronNetworkType: 'geneve,vlan'
  NeutronNetworkVLANRanges: 'leaf0:1:1000,leaf1:1:1000,leaf2:1:1000'

  Controller0Parameters:
    NeutronBridgeMappings: "leaf0:br-ex"

  Controller1Parameters:
    NeutronBridgeMappings: "leaf0:br-ex"

  Controller2Parameters:
    NeutronBridgeMappings: "leaf0:br-ex"

  Compute0Parameters:
    NeutronBridgeMappings: "leaf0:br-ex"

  Compute1Parameters:
    NeutronBridgeMappings: "leaf1:br-ex"

  Compute2Parameters:
    NeutronBridgeMappings: "leaf2:br-ex"

      Note

      You can use both flat networks and VLANs in your spine-leaf topology.

  4. Add the control plane subnet mapping for each spine-leaf network by using the <role>ControlPlaneSubnet parameter:

    Example

    parameter_defaults:
  NeutronFlatNetworks: leaf0,leaf1,leaf2
  Controller0Parameters:
    NeutronBridgeMappings: "leaf0:br-ex"
    ControllerControlPlaneSubnet: leaf0
  Controller1Parameters:
    NeutronBridgeMappings: "leaf0:br-ex"
    Controller1ControlPlaneSubnet: leaf0
  Controller2Parameters:
    NeutronBridgeMappings: "leaf0:br-ex"
    Controller2ControlPlaneSubnet: leaf0
  Compute0Parameters:
    NeutronBridgeMappings: "leaf0:br-ex"
    Compute0ControlPlaneSubnet: leaf0
  CephStorage0Parameters:
    CephStorage0ControlPlaneSubnet: leaf0
  Compute1Parameters:
    NeutronBridgeMappings: "leaf1:br-ex"
    Compute1ControlPlaneSubnet: leaf1
  CephStorage1Parameters:
    CephStorage1ControlPlaneSubnet: leaf1
  Compute2Parameters:
    NeutronBridgeMappings: "leaf2:br-ex"
    Compute2ControlPlaneSubnet: leaf2
  CephStorage2Parameters:
    CephStorage2ControlPlaneSubnet: leaf2

Next steps

  1. Note the path and file name of the custom network environment file that you have created. You will need this information later when you deploy your overcloud.
  2. Proceed to the next step Setting the subnet for virtual IP addresses.

Additional resources

4.5. Setting the subnet for virtual IP addresses

By default, the Red Hat Openstack Platform (RHOSP) Controller role hosts virtual IP (VIP) addresses for each network. The RHOSP overcloud takes the VIPs from the base subnet of each network except for the control plane. The control plane uses ctlplane-subnet, which is the default subnet name created during a standard undercloud installation.

In the spine-leaf examples used in this document, the default base provisioning network is leaf0 instead of ctlplane-subnet. This means that you must add the value pair subnet: leaf0 to the network:ctlplane parameter to map the subnet to leaf0.

Complete the following steps to create a YAML-formatted, custom network VIP definition file that contains the overrides for your VIPs on the overcloud. Later, the provisioning process creates a heat environment file from your network VIP definition file that you include when you deploy your RHOSP overcloud.

Prerequisites

  • Access to the undercloud host and credentials for the stack user.

Procedure

  1. Log in to the undercloud host as the stack user.

  2. Source the stackrc undercloud credentials file: 

    $ source ~/stackrc

  3. In a new custom network VIP definition template, such as spine-leaf-vip-data.yaml, list the virtual IP addresses that need to be created on the specific subnet used by controller nodes.

    Example

    - network: storage_mgmt
  subnet: storage_mgmt_subnet_leaf1
- network: internal_api
  subnet: internal_api_subnet_leaf1
- network: storage
  subnet: storage_subnet_leaf1
- network: external
  subnet: external_subnet_leaf1
  ip_address: 172.20.11.50
- network: ctlplane
  subnet: leaf0
- network: oc_provisioning
  subnet: oc_provisioning_subnet_leaf1
- network: storage_nfs
  subnet: storage_nfs_subnet_leaf1

    You can use the following parameters in your spine-leaf-vip-data.yaml file:

    network
    Sets the neutron network name. This is the only required parameter.
    ip_address
    Sets the IP address of the VIP.
    subnet
    Sets the neutron subnet name. Use to specify the subnet when creating the virtual IP neutron port. This parameter is required when your deployment uses routed networks.
    dns_name
    Sets the FQDN (Fully Qualified Domain Name).
    name

    Sets the virtual IP name.

    Tip

    For more information, see Adding a composable network in the Installing and managing Red Hat OpenStack Platform with director guide.

Next steps

  1. Note the path and file name of the custom network VIP definition template that you have created. You use this path later when you provision your network VIPs for the RHOSP overcloud.
  2. Proceed to the next step Provisioning networks and VIPs for the overcloud.

4.6. Provisioning networks and VIPs for the overcloud

The Red Hat OpenStack Platform (RHOSP) provisioning process uses your network definition file to create a new heat environment file that contains your network specifications. If your deployment uses VIPs, RHOSP creates a new heat environment file from your VIP definition file. After you provision your networks and VIPs, you have two heat environment files that you use later to deploy your overcloud.

Prerequisites

  • Access to the undercloud host and credentials for the stack user.
  • You have a network configuration template.
  • If you are using VIPs, you have a VIP definition template.

Procedure

  1. Log in to the undercloud host as the stack user.

  2. Source the stackrc undercloud credentials file: 

    $ source ~/stackrc

  3. Provision your overcloud networks.

    Use the overcloud network provision command, and provide the path to the network definition file that you created earlier.

    Tip

    For more information, see Configuring and provisioning overcloud network definitions in the Installing and managing Red Hat OpenStack Platform with director guide.

    Example

    In this example, the path is /home/stack/templates/spine-leaf-networks-data.yaml. Use the --output argument to name the file created by the command. 

    $ openstack overcloud network provision \
  --output spine-leaf-networks-provisioned.yaml \
 /home/stack/templates/spine-leaf-networks-data.yaml
    Important

    The name of the output file that you specify must end in either .yaml or .template.

  4. Provision your overcloud VIPs.

    Use the overcloud network vip provision command, with the --stack argument to name the VIP definition file that you created earlier. Use the --output argument to name the file created by the command.

    Tip

    For more information, see Configuring and provisioning network VIPs for the overcloud in the Installing and managing Red Hat OpenStack Platform with director guide. 

    $ openstack overcloud network vip provision \
  --stack spine-leaf-overcloud \
 --output spine-leaf-vips-provisioned.yaml \
 /home/stack/templates/spine-leaf-vip-data.yaml
    Important

    The name of the output file that you specify must end in either .yaml or .template.

  5. Note the path and file names of the generated output files. You use this information later when you deploy your overcloud.

Verification

  • You can use the following commands to confirm that the command created your overcloud networks and subnets: 

    $ openstack network list
$ openstack subnet list
$ openstack network show <network>
$ openstack subnet show <subnet>
$ openstack port list
$ openstack port show <port>

    Replace <network>, <subnet>, and <port> with the name or UUID of the network, subnet, and port that you want to check.

Next steps

  1. If you are using pre-provisioned nodes, skip to Running the overcloud deployment command.
  2. Otherwise, proceed to the next step Registering bare metal nodes on the overcloud.

Additional resources

4.7. Registering bare metal nodes on the overcloud

Red Hat OpenStack Platform (RHOSP) director requires a custom node definition template that specifies the hardware and power management details of your physical machines. You can create this template in JSON or YAML formats. After you register your physical machines as bare metal nodes, you introspect them, and then you finally provision them.

Note

If you are using pre-provisioned bare metal nodes then you can skip registering, introspecting, and provisioning bare metal nodes, and go to Deploying a spine-leaf enabled overcloud.

Prerequisites

  • Access to the undercloud host and credentials for the stack user.

Procedure

  1. Log in to the undercloud host as the stack user.

  2. Source the stackrc undercloud credentials file: 

    $ source ~/stackrc

  3. Create a new node definition template, such as barematal-nodes.yaml. Add a list of your physical machines that includes their hardware and power management details.

    Example

    nodes:
  - name: "node01"
    ports:
      - address: "aa:aa:aa:aa:aa:aa"
        physical_network: ctlplane
        local_link_connection:
          switch_id: 52:54:00:00:00:00
          port_id: p0
    cpu: 4
    memory: 6144
    disk: 40
    arch: "x86_64"
    pm_type: "ipmi"
    pm_user: "admin"
    pm_password: "p@55w0rd!"
    pm_addr: "192.168.24.205"
  - name: "node02"
    ports:
      - address: "bb:bb:bb:bb:bb:bb"
        physical_network: ctlplane
        local_link_connection:
          switch_id: 52:54:00:00:00:00
          port_id: p0
    cpu: 4
    memory: 6144
    disk: 40
    arch: "x86_64"
    pm_type: "ipmi"
    pm_user: "admin"
    pm_password: "p@55w0rd!"
    pm_addr: "192.168.24.206"

    Tip

    For more information about template parameter values and for a JSON example, see Registering nodes for the overcloud in the Installing and managing Red Hat OpenStack Platform with director guide.

  4. Verify the template formatting and syntax.

    Example

    $ openstack overcloud node import --validate-only ~/templates/\
baremetal-nodes.yaml

  5. Correct any errors and save your node definition template.

  6. Import your node definition template to RHOSP director to register each node from your template into director:

    Example

    $ openstack overcloud node import ~/baremetal-nodes.yaml

Verification

  • When the node registration and configuration is complete, confirm that director has successfully registered the nodes: 

    $ openstack baremetal node list

    The baremetal node list command should include the imported nodes and the status should be manageable.

Next steps

Additional resources

4.8. Introspecting bare metal nodes on the overcloud

After you register a physical machine as a bare metal node, you can use OpenStack Platform (RHOSP) director introspection to automatically add then node’s hardware details and create ports for each of its Ethernet MAC addresses. After you perform introspection on your bare metal nodes, the final step is to provision them.

Note

If you are using pre-provisioned bare metal nodes then you can skip introspecting and introspecting bare metal nodes and go to Deploying a spine-leaf enabled overcloud.

Prerequisites

  • Access to the undercloud host and credentials for the stack user.
  • You have registered your bare metal nodes for your overcloud with RHOSP.

Procedure

  1. Log in to the undercloud host as the stack user.

  2. Source the stackrc undercloud credentials file: 

    $ source ~/stackrc

  3. Run the pre-introspection validation group to check the introspection requirements: 

    $ validation run --group pre-introspection
  4. Review the results of the validation report.

  5. (Optional) Review detailed output from a specific validation: 

    $ validation history get --full <UUID>

    Replace <UUID> with the UUID of the specific validation from the report that you want to review.

    Important

    A FAILED validation does not prevent you from deploying or running RHOSP. However, a FAILED validation can indicate a potential issue with a production environment.

  6. Inspect the hardware attributes of all nodes: 

    $ openstack overcloud node introspect --all-manageable --provide
    Tip

    For more information, see Using director introspection to collect bare metal node hardware information in the Installing and managing Red Hat OpenStack Platform with director guide.

    Monitor the introspection progress logs in a separate terminal window: 

    $ sudo tail -f /var/log/containers/ironic-inspector/ironic-inspector.log

Verification

  • After the introspection completes, all nodes change to an available state.

Next steps

Additional resources

4.9. Provisioning bare metal nodes for the overcloud

To provision your bare metal nodes for Red Hat OpenStack Platform (RHOSP), you define the quantity and attributes of the bare metal nodes that you want to deploy and assign overcloud roles to these nodes. You also define the network layout of the nodes. You add all of this information in a node definition file in YAML format.

The provisioning process creates a heat environment file from your node definition file. This heat environment file contains the node specifications you configured in your node definition file, including node count, predictive node placement, custom images, and custom NICs. When you deploy your overcloud, include this heat environment file in the deployment command. The provisioning process also provisions the port resources for all networks defined for each node or role in the node definition file.

Note

If you are using pre-provisioned bare metal nodes then you can skip provisioning bare metal nodes and go to Deploying a spine-leaf enabled overcloud.

Prerequisites

  • Access to the undercloud host and credentials for the stack user.
  • The bare metal nodes are registered, introspected, and available for provisioning and deployment.

Procedure

  1. Log in to the undercloud host as the stack user.

  2. Source the stackrc undercloud credentials file: 

    $ source ~/stackrc

  3. Create a bare metal nodes definition file, such as spine-leaf-baremetal-nodes.yaml, and define the node count for each role that you want to provision.

    Example

    - name: Controller
  count: 3
  defaults:
    networks:
    - network: ctlplane
      vif: true
    - network: external
      subnet: external_subnet
    - network: internal_api
      subnet: internal_api_subnet01
    - network: storage
      subnet: storage_subnet01
    - network: storage_mgmt
      subnet: storage_mgmt_subnet01
    - network: tenant
      subnet: tenant_subnet01
    network_config:
      template: /home/stack/templates/spine-leaf-nics/single-nic-vlans.j2
      default_route_network:
      - external
- name: Compute0
  count: 1
  defaults:
    networks:
    - network: ctlplane
      vif: true
    - network: internal_api
      subnet: internal_api_subnet02
    - network: tenant
      subnet: tenant_subnet02
    - network: storage
      subnet: storage_subnet02
    network_config:
      template: /home/stack/templates/spine-leaf-nics/single-nic-vlans.j2
- name: Compute1
...

    Tip

    For more information about the properties that you can set bare metal node definition file, see Provisioning bare metal nodes for the overcloud in the Installing and managing Red Hat OpenStack Platform with director guide.

  4. Provision the overcloud bare metal nodes, using the overcloud node provision command.

    Example

    $ openstack overcloud node provision \
 --stack spine_leaf_overcloud \
 --network-config \
 --output spine-leaf-baremetal-nodes-provisioned.yaml \
 /home/stack/templates/spine-leaf-baremetal-nodes.yaml

    Important

    The name of the output file that you specify must end in either .yaml or .template.

  5. Monitor the provisioning progress in a separate terminal. When provisioning is successful, the node state changes from available to active: 

    $ watch openstack baremetal node list

  6. Use the metalsmith tool to obtain a unified view of your nodes, including allocations and ports: 

    $ metalsmith list
  7. Note the path and file name of the generated output file. You need this path later when you deploy your overcloud.

Verification

  • Confirm the association of nodes to hostnames: 

    $ openstack baremetal allocation list

Next steps

Additional resources

4.10. Deploying a spine-leaf enabled overcloud

The last step in deploying your Red Hat OpenStack Platform (RHOSP) overcloud is to run the overcloud deploy command. Inputs to the command include all of the various overcloud templates and environment files that you constructed. RHOSP director uses these templates and files as a plan for how to install and configure your overcloud.

Prerequisites

  • Access to the undercloud host and credentials for the stack user.
  • You have performed all of the steps listed in the earlier procedures in this section and have assembled all of the various heat templates and environment files to use as inputs for the overcloud deploy command.

Procedure

  1. Log in to the undercloud host as the stack user.

  2. Source the stackrc undercloud credentials file: 

    $ source ~/stackrc

  3. Collate the custom environment files and custom templates that you need for your overcloud environment. This list includes the unedited heat template files provided with your director installation and the custom files you created. Ensure that you have the paths to the following files:

    • Your custom network definition file that contains the specifications for your spine-leaf network on the overcloud, for example, spine-leaf-networks-data.yaml.

      For more information, see Defining the leaf networks.

    • Your custom roles data file that defines a role for each leaf.

      Example: spine-leaf-roles.yaml.

      For more information, see Defining leaf roles and attaching networks

    • Your custom environment file that contains the roles and the custom NIC template mappings for each role.

      Example: spine-leaf-nic-roles-map.yaml.

      For more information, see Creating a custom NIC configuration for leaf roles.

    • Your custom network environment file that contains the separate network mappings and sets access to the control plane networks for the overcloud.

      Example: spine-leaf-ctlplane.yaml

      For more information, see Configuring the leaf networks.

    • The output file from provisioning your overcloud networks.

      Example: spine-leaf-networks-provisioned.yaml

      For more information, see Provisioning networks and VIPs for the overcloud.

    • The output file from provisioning your overcloud VIPs.

      Example: spine-leaf-vips-provisioned.yaml

      For more information, see Provisioning networks and VIPs for the overcloud.

    • If you are not using pre-provisioned nodes, the output file from provisioning bare-metal nodes.

      Example: spine-leaf-baremetal-nodes-provisioned.yaml.

      For more information, see Provisioning bare metal nodes for the overcloud.

    • Any other custom environment files.

  4. Enter the overcloud deploy command by carefully ordering the custom environment files and custom templates that are inputs to the command.

    The general rule is to specify any unedited heat template files first, followed by your custom environment files and custom templates that contain custom configurations, such as overrides to the default properties.

    Follow this order for listing the inputs to the overcloud deploy command:

    1. Include your custom environment file that contains your custom NIC templates mapped to each role.

      Example: spine-leaf-nic-roles-map.yaml, after network-environment.yaml.

      The network-environment.yaml file provides the default network configuration for composable network parameters, that your mapping file overrides. Note that the director renders this file from the network-environment.j2.yaml Jinja2 template.

    2. If you created any other spine leaf network environment files, include these environment files after the roles-NIC templates mapping file.

    3. Add any additional environment files. For example, an environment file with your container image locations or Ceph cluster configuration.

      Example

      This excerpt from a sample overcloud deploy command demonstrates the proper ordering of the command’s inputs: 

      $ openstack overcloud deploy --templates \
  -n /home/stack/templates/spine-leaf-networks-data.yaml \
  -e /usr/share/openstack-tripleo-heat-templates/environments/network-environment.yaml \
  -e /usr/share/openstack-tripleo-heat-templates/environments/services/frr.yaml \
  -e /usr/share/openstack-tripleo-heat-templates/environments/services/ovn-bgp-agent.yaml \
  -e /home/stack/templates/spine-leaf-nic-roles-map.yaml \
  -e /home/stack/templates/spine-leaf-ctlplane.yaml \
  -e /home/stack/templates/spine-leaf-baremetal-provisioned.yaml \
  -e /home/stack/templates/spine-leaf-networks-provisioned.yaml \
  -e /home/stack/templates/spine-leaf-vips-provisioned.yaml \
  -e /home/stack/containers-prepare-parameter.yaml \
  -e /home/stack/inject-trust-anchor-hiera.yaml \
  -r /home/stack/templates/spine-leaf-roles-data.yaml
  ...
      Tip

      For more information, see Creating your overcloud in the Installing and managing Red Hat OpenStack Platform with director guide.

  5. Run the overcloud deploy command.

    When the overcloud creation is finished, the RHOSP director provides details to help you access your overcloud.

Verification

Additional resources

4.11. Adding a new leaf to a spine-leaf deployment

When increasing network capacity or adding a new physical site, you might need to add a new leaf to your Red Hat OpenStack Platform (RHOSP) spine-leaf network.

Prerequisites

  • Your RHOSP deployment uses a spine-leaf network topology.

Procedure

  1. Log in to the undercloud host as the stack user.

  2. Source the stackrc undercloud credentials file: 

    $ source ~/stackrc

  3. Open your network definition template, for example, /home/stack/templates/spine-leaf-networks-data.yaml. Under the appropriate base network, add a leaf subnet as a composable network item for the new leaf that you are adding.

    Example

    In this example, a subnet entry for the new leaf (leaf3) has been added: 

    - name: InternalApi
  name_lower: internal_api
  vip: true
  vlan: 10
  ip_subnet: '172.18.0.0/24'
  allocation_pools: [{'start': '172.18.0.4', 'end': '172.18.0.250'}]
  gateway_ip: '172.18.0.1'
  subnets:
    internal_api_leaf1:
      vlan: 11
      ip_subnet: '172.18.1.0/24'
      allocation_pools: [{'start': '172.18.1.4', 'end': '172.18.1.250'}]
      gateway_ip: '172.18.1.1'
    internal_api_leaf2:
      vlan: 12
      ip_subnet: '172.18.2.0/24'
      allocation_pools: [{'start': '172.18.2.4', 'end': '172.18.2.250'}]
      gateway_ip: '172.18.2.1'
    internal_api_leaf3:
      vlan: 13
      ip_subnet: '172.18.3.0/24'
      allocation_pools: [{'start': '172.18.3.4', 'end': '172.18.3.250'}]
      gateway_ip: '172.18.3.1'

  4. Create a roles data file for the new leaf that you are adding.

    1. Copy a leaf Compute and a leaf Ceph Storage file for the new leaf that you are adding.

      Example

      In this example, Compute1.yaml and CephStorage1.yaml are copied for the new leaf, Compute3.yaml and CephStorage3.yaml, respectively: 

      $ cp ~/roles/Compute1.yaml ~/roles/Compute3.yaml
$ cp ~/roles/CephStorage1.yaml ~/roles/CephStorage3.yaml

    2. Edit the name and HostnameFormatDefault parameters in the new leaf files so that they align with the respective leaf parameters.

      Example

      For example, the parameters in the Leaf 1 Compute file have the following values: 

      - name: ComputeLeaf1
  HostnameFormatDefault: '%stackname%-compute-leaf1-%index%'

      Example

      The Leaf 1 Ceph Storage parameters have the following values: 

      - name: CephStorageLeaf1
  HostnameFormatDefault: '%stackname%-cephstorage-leaf1-%index%'

    3. Edit the network parameter in the new leaf files so that they align with the respective Leaf network parameters.

      Example

      For example, the parameters in the Leaf 1 Compute file have the following values: 

      - name: ComputeLeaf1
  networks:
    InternalApi:
      subnet: internal_api_leaf1
    Tenant:
      subnet: tenant_leaf1
    Storage:
      subnet: storage_leaf1

      Example

      The Leaf 1 Ceph Storage parameters have the following values: 

      - name: CephStorageLeaf1
  networks:
    Storage:
      subnet: storage_leaf1
    StorageMgmt:
      subnet: storage_mgmt_leaf1

    4. When your role configuration is complete, run the following command to generate the full roles data file. Include all of the leafs in your network and the new leaf that you are adding.

      Example

      In this example, leaf3 is added to leaf0, leaf1, and leaf2: 

      $ openstack overcloud roles generate --roles-path ~/roles -o roles_data_spine_leaf.yaml Controller Controller1 Controller2 Compute Compute1 Compute2 Compute3 CephStorage CephStorage1 CephStorage2 CephStorage3

      This creates a full roles_data_spine_leaf.yaml file that includes all of the custom roles for each respective leaf network.

  5. Create a custom NIC configuration for the leaf that you are adding.

    1. Copy a leaf Compute and a leaf Ceph Storage NIC configuration file for the new leaf that you are adding.

      Example

      In this example, computeleaf1.yaml and ceph-storageleaf1.yaml are copied for the new leaf, computeleaf3.yaml and ceph-storageleaf3.yaml, respectively: 

      $ cp ~/templates/spine-leaf-nics/computeleaf1.yaml ~/templates/spine-leaf-nics/computeleaf3.yaml
$ cp ~/templates/spine-leaf-nics/ceph-storageleaf1.yaml ~/templates/spine-leaf-nics/ceph-storageleaf3.yaml

  6. Open your custom environment file that contains the roles and the custom NIC template mappings for each role, for example, spine-leaf-nic-roles-map.yaml. Insert an entry for each role for the new leaf that you are adding. 

    parameter_defaults:
  %%ROLE%%NetworkConfigTemplate: <path_to_ansible_jinja2_nic_config_file>

    Example

    In this example, the entries ComputeLeaf3NetworkConfigTemplate and CephStorage3NetworkConfigTemplate have been added: 

    parameter_defaults:
  Controller0NetworkConfigTemplate: '/home/stack/templates/spine-leaf-nics/single-nic-vlans.j2'
  Controller1NetworkConfigTemplate: '/home/stack/templates/spine-leaf-nics/single-nic-vlans.j2'
  Controller2NetworkConfigTemplate: '/home/stack/templates/spine-leaf-nics/single-nic-vlans.j2'
  ComputeLeaf0NetworkConfigTemplate: '/home/stack/templates/spine-leaf-nics/single-nic-vlans.j2'
  ComputeLeaf1NetworkConfigTemplate: '/home/stack/templates/spine-leaf-nics/single-nic-vlans.j2'
  ComputeLeaf2NetworkConfigTemplate: '/home/stack/templates/spine-leaf-nics/single-nic-vlans.j2'
  ComputeLeaf3NetworkConfigTemplate: '/home/stack/templates/spine-leaf-nics/single-nic-vlans.j2'
  CephStorage0NetworkConfigTemplate: '/home/stack/templates/spine-leaf-nics/single-nic-vlans.j2'
  CephStorage1NetworkConfigTemplate: '/home/stack/templates/spine-leaf-nics/single-nic-vlans.j2'
  CephStorage2NetworkConfigTemplate: '/home/stack/templates/spine-leaf-nics/single-nic-vlans.j2'
  CephStorage3NetworkConfigTemplate: '/home/stack/templates/spine-leaf-nics/single-nic-vlans.j2'

  7. Open your custom network environment file that contains the separate network mappings and sets access to the control plane networks for the overcloud, for example, spine-leaf-ctlplane.yaml and update the control plane parameters.

    Under the parameter_defaults section, add the control plane subnet mapping for the new leaf network. Also, include the external network mapping for the new leaf network.

    • For flat network mappings, list the new leaf (leaf3) in the NeutronFlatNetworks parameter and set the NeutronBridgeMappings parameter for the new leaf: 

      parameter_defaults:
  NeutronFlatNetworks: leaf0,leaf1,leaf2,leaf3
  Controller0Parameters:
    NeutronBridgeMappings: "leaf0:br-ex"
  Compute0Parameters:
    NeutronBridgeMappings: "leaf0:br-ex"
  Compute1Parameters:
    NeutronBridgeMappings: "leaf1:br-ex"
  Compute2Parameters:
    NeutronBridgeMappings: "leaf2:br-ex"
  Compute3Parameters:
    NeutronBridgeMappings: "leaf3:br-ex"

    • For VLAN network mappings, additionally set the NeutronNetworkVLANRanges to map VLANs for the new leaf (leaf3) network: 

        NeutronNetworkType: 'geneve,vlan'
  NeutronNetworkVLANRanges: 'leaf0:1:1000,leaf1:1:1000,leaf2:1:1000,leaf3:1:1000'

      Example

      In this example, flat network mappings are used, and the new leaf (leaf3) entries are added: 

      parameter_defaults:
  NeutronFlatNetworks: leaf0,leaf1,leaf2,leaf3
  Controller0Parameters:
    NeutronBridgeMappings: "leaf0:br-ex"
  ControllerControlPlaneSubnet: leaf0
  Controller1Parameters:
    NeutronBridgeMappings: "leaf0:br-ex"
  Controller1ControlPlaneSubnet: leaf0
  Controller2Parameters:
    NeutronBridgeMappings: "leaf0:br-ex"
  Controller2ControlPlaneSubnet: leaf0
  Compute0Parameters:
    NeutronBridgeMappings: "leaf0:br-ex"
  Compute0ControlPlaneSubnet: leaf0
  Compute1Parameters:
    NeutronBridgeMappings: "leaf1:br-ex"
  Compute1ControlPlaneSubnet: leaf1
  Compute2Parameters:
    NeutronBridgeMappings: "leaf2:br-ex"
  Compute2ControlPlaneSubnet: leaf2
  Compute3Parameters:
    NeutronBridgeMappings: "leaf3:br-ex"
  Compute3ControlPlaneSubnet: leaf3

  8. Provision your modified network.

    For more information, see Provision your overcloud networks and overcloud VIPs.

  9. In your bare metal nodes definition file that you created earlier, for example, spine-leaf-baremetal-nodes.yaml, ensure that the network_config_update variable is set to true.

    Example

    - name: Controller
  count: 3
  defaults:
    networks:
    - network: ctlplane
      vif: true
    - network: external
      subnet: external_subnet
    - network: internal_api
      subnet: internal_api_subnet01
    - network: storage
      subnet: storage_subnet01
    - network: storage_mgmt
      subnet: storage_mgmt_subnet01
    - network: tenant
      subnet: tenant_subnet01
    network_config:
      template: /home/stack/templates/spine-leaf-nics/single-nic-vlans.j2
      default_route_network:
      - external
    network_config_update: true

  10. Provision your modified nodes.

    For more information, see Provision bare metal nodes.

  11. Redeploy your spine-leaf enabled overcloud, by following the steps in Deploying a spine-leaf enabled overcloud.