Basic Deployment (CLI)

These steps document a basic deployment with TripleO in an environment using the project defaults.

Note

Since Rocky, Ansible is used to deploy the software configuration of the overcloud nodes using a feature called config-download. While there are no necessary changes to the default deployment commands, there are several differences to the deployer experience.

It’s recommended to review these differences as documented at Ansible config-download differences

config-download is fully documented at TripleO config-download User’s Guide: Deploying with Ansible

Prepare Your Environment

  1. Make sure you have your environment ready and undercloud running:

  2. Log into your undercloud virtual machine and become the non-root user (stack by default):

    ssh root@<undercloud-machine>
    
    su - stack
    
  3. In order to use CLI commands easily you need to source needed environment variables:

    source stackrc
    

Get Images

Note

If you already have images built, perhaps from a previous installation of TripleO, you can simply copy those image files into your non-root user’s home directory and skip this section.

If you do this, be aware that sometimes newer versions of TripleO do not work with older images, so if the deployment fails it may be necessary to delete the older images and restart the process from this step.

Alternatively, images are available via RDO at https://images.rdoproject.org/centos8/master/rdo_trunk/ which offers images from both the CentOS Build System (cbs) and RDO Trunk (called rdo_trunk or delorean). However this mirror is slow so if you experience slow download speeds you should skip to building the images instead.

The image files required are:

ironic-python-agent.initramfs
ironic-python-agent.kernel
overcloud-full.initrd
overcloud-full.qcow2
overcloud-full.vmlinuz

Images must be built prior to doing a deployment. An IPA ramdisk and openstack-full image can all be built using tripleo-common.

It’s recommended to build images on the installed undercloud directly since all the dependencies are already present, but this is not a requirement.

The following steps can be used to build images. They should be run as the same non-root user that was used to install the undercloud. If the images are not created on the undercloud, one should use a non-root user.

  1. Choose image operating system:

    CentOS

    The image build with no arguments will build CentOS 7. It will include the common YAML of /usr/share/openstack-tripleo-common/image-yaml/overcloud-images.yaml and the CentOS YAML at /usr/share/openstack-tripleo-common/image-yaml/overcloud-images-centos7.yaml.

    CentOS 8

    The default YAML for Centos 8 is

    /usr/share/openstack-tripleo-common/image-yaml/overcloud-images-centos8.yaml

    export OS_YAML="/usr/share/openstack-tripleo-common/image-yaml/overcloud-images-centos8.yaml"
    

    RHEL

    The common YAML is /usr/share/openstack-tripleo-common/image-yaml/overcloud-images.yaml. It must be specified along with the following.

    The default YAML for RHEL is /usr/share/openstack-tripleo-common/image-yaml/overcloud-images-rhel7.yaml

    export OS_YAML="/usr/share/openstack-tripleo-common/image-yaml/overcloud-images-rhel7.yaml"
    
  2. Install the current-tripleo delorean repository and deps repository:

    Note

    Python3 is required for Ussuri and newer releases of OpenStack which is supported on RHEL 8 and CentOS 8. Train is also recommended to be installed on RHEL 8 or CentOS 8.

    1. Download and install the python-tripleo-repos RPM from the appropriate RDO repository

      CentOS 8 and CentOS Strem 8

      Current Centos 8 RDO repository.

      sudo dnf install -y https://trunk.rdoproject.org/centos8/component/tripleo/current/python3-tripleo-repos-<version>.el8.noarch.rpm
      

      Note

      tripleo-repos removes any repositories that it manages before each run. This means all repositories must be specified in a single tripleo-repos call. As an example, the correct way to install the current and ceph repos is to run tripleo-repos current ceph, not two separate calls.

      Stable Branch

      Enable the appropriate repos for the desired release, as indicated below. Do not enable any other repos not explicitly marked for that release.

      Wallaby

      Enable the current Wallaby repositories

      sudo -E tripleo-repos -b wallaby current
      

      Ceph

      Include the Ceph repo in the tripleo-repos call

      sudo -E tripleo-repos -b wallaby current ceph
      

      Victoria

      Enable the current Victoria repositories

      sudo -E tripleo-repos -b victoria current
      

      Ceph

      Include the Ceph repo in the tripleo-repos call

      sudo -E tripleo-repos -b victoria current ceph
      

      Ussuri

      Enable the current Ussuri repositories

      sudo -E tripleo-repos -b ussuri current
      

      Ceph

      Include the Ceph repo in the tripleo-repos call

      sudo -E tripleo-repos -b ussuri current ceph
      

      Train

      Enable the current Train repositories

      sudo -E tripleo-repos -b train current
      

      Ceph

      Include the Ceph repo in the tripleo-repos call

      sudo -E tripleo-repos -b train current ceph
      

    Warning

    The remaining repositories configuration steps below should not be done for stable releases!

    1. Run tripleo-repos to install the appropriate repositories. The option below will enable the latest master TripleO packages, the latest promoted packages for all other OpenStack services and dependencies and the latest stable Ceph packages. There are other repository configurations available in tripleo-repos, see its --help output for details.

      sudo -E tripleo-repos current-tripleo-dev ceph
      
  1. Export environment variables

    export DIB_YUM_REPO_CONF="/etc/yum.repos.d/delorean*"
    

    Ceph

    export DIB_YUM_REPO_CONF="$DIB_YUM_REPO_CONF /etc/yum.repos.d/tripleo-centos-ceph*.repo"
    

    CentOS 8

    export DIB_YUM_REPO_CONF="/etc/yum.repos.d/delorean* /etc/yum.repos.d/tripleo-centos-*"
    

    Stable Branch

    Victoria

    export STABLE_RELEASE="victoria"
    

    Ussuri

    export STABLE_RELEASE="ussuri"
    

    Train

    export STABLE_RELEASE="train"
    
  2. Build the required images:

    RHEL

    Download the RHEL 7.4 cloud image or copy it over from a different location, for example: https://access.redhat.com/downloads/content/69/ver=/rhel---7/7.4/x86_64/product-software, and define the needed environment variables for RHEL 7.4 prior to running tripleo-build-images:

    export DIB_LOCAL_IMAGE=rhel-server-7.4-x86_64-kvm.qcow2
    

    RHEL Portal Registration

    To register the image builds to the Red Hat Portal define the following variables:

    export REG_METHOD=portal
    export REG_USER="[your username]"
    export REG_PASSWORD="[your password]"
    # Find this with `sudo subscription-manager list --available`
    export REG_POOL_ID="[pool id]"
    export REG_REPOS="rhel-7-server-rpms rhel-7-server-extras-rpms rhel-ha-for-rhel-7-server-rpms \
        rhel-7-server-optional-rpms rhel-7-server-openstack-6.0-rpms"
    

    Ceph

    If using Ceph, additional channels need to be added to REG_REPOS. Enable the appropriate channels for the desired release, as indicated below. Do not enable any other channels not explicitly marked for that release.

    rhel-7-server-rhceph-2-mon-rpms
    rhel-7-server-rhceph-2-osd-rpms
    rhel-7-server-rhceph-2-tools-rpms
    

    RHEL Satellite Registration

    To register the image builds to a Satellite define the following variables. Only using an activation key is supported when registering to Satellite, username/password is not supported for security reasons. The activation key must enable the repos shown:

    export REG_METHOD=satellite
    # REG_SAT_URL should be in the format of:
    # http://<satellite-hostname>
    export REG_SAT_URL="[satellite url]"
    export REG_ORG="[satellite org]"
    # Activation key must enable these repos:
    # rhel-7-server-rpms
    # rhel-7-server-optional-rpms
    # rhel-7-server-extras-rpms
    # rhel-7-server-openstack-6.0-rpms
    # rhel-7-server-rhceph-{2,1.3}-mon-rpms
    # rhel-7-server-rhceph-{2,1.3}-osd-rpms
    # rhel-7-server-rhceph-{2,1.3}-tools-rpms
    export REG_ACTIVATION_KEY="[activation key]"
    
    openstack overcloud image build
    

    RHEL

    openstack overcloud image build --config-file /usr/share/openstack-tripleo-common/image-yaml/overcloud-images.yaml --config-file $OS_YAML
    

    CentOS 8

    openstack overcloud image build --config-file /usr/share/openstack-tripleo-common/image-yaml/overcloud-images-python3.yaml --config-file $OS_YAML
    

    See the help for openstack overcloud image build for further options.

    The YAML files are cumulative. Order on the command line is important. The packages, elements, and options sections will append. All others will overwrite previously read values.

    Note

    This command will build overcloud-full images (*.qcow2, *.initrd, *.vmlinuz) and ironic-python-agent images (*.initramfs, *.kernel)

    In order to build specific images, one can use the --image-name flag to openstack overcloud image build. It can be specified multiple times.

Note

If you want to use whole disk images with TripleO, please see Use whole disk images for overcloud.

Upload Images

Load the images into the containerized undercloud Glance:

openstack overcloud image upload

To upload a single image, see Uploading a Single Image.

If working with multiple architectures and/or platforms with an architecture these attributes can be specified at upload time as in:

openstack overcloud image upload
openstack overcloud image upload --arch x86_64 \
    --httpboot /var/lib/ironic/httpboot/x86_64
openstack overcloud image upload --arch x86_64 --platform SNB \
    --httpboot /var/lib/ironic/httpboot/x86_64-SNB

Note

Adding --httpboot is optional but suggested if you need to ensure that the agent images are unique within your environment.

Prior to Rocky release

Before Rocky, the undercloud isn’t containerized by default. Hence you should use the /httpboot/* paths instead.

This will create 3 sets of images with in the undercloud image service for later use in deployment, see Baremetal Environment

Register Nodes

Register and configure nodes for your deployment with Ironic:

openstack overcloud node import instackenv.json

The file to be imported may be either JSON, YAML or CSV format, and the type is detected via the file extension (json, yaml, csv). The file format is documented in instackenv.json.

The nodes status will be set to manageable by default, so that introspection may later be run. To also run introspection and make the nodes available for deployment in one step, the following flags can be used:

openstack overcloud node import --introspect --provide instackenv.json

Starting with the Newton release you can take advantage of the enroll provisioning state - see Bare Metal Node States for details.

If your hardware has several hard drives, it’s highly recommended that you specify the exact device to be used during introspection and deployment as a root device. Please see Setting the Root Device for Deployment for details.

Warning

If you don’t specify the root device explicitly, any device may be picked. Also the device chosen automatically is NOT guaranteed to be the same across rebuilds. Make sure to wipe the previous installation before rebuilding in this case.

If there is information from previous deployments on the nodes’ disks, it is recommended to at least remove the partitions and partition table(s). See Node cleaning for information on how to do it.

Finally, if you want your nodes to boot in the UEFI mode, additional steps may have to be taken - see Booting in UEFI mode for details.

Warning

It’s not recommended to delete nodes and/or rerun this command after you have proceeded to the next steps. Particularly, if you start introspection and then re-register nodes, you won’t be able to retry introspection until the previous one times out (1 hour by default). If you are having issues with nodes after registration, please follow Node Registration and Management Problems.

Another approach to enrolling node is Node Discovery.

Introspect Nodes

Validations

Once the undercloud is installed, you can run the pre-introspection validations:

openstack tripleo validator run --group pre-introspection

Then verify the results as described in Running validation groups.

Nodes must be in the manageable provisioning state in order to run introspection. Introspect hardware attributes of nodes with:

openstack overcloud node introspect --all-manageable

Nodes can also be specified individually by UUID. The --provide flag can be used in order to move the nodes automatically to the available provisioning state once the introspection is finished, making the nodes available for deployment.

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

Note

Introspection has to finish without errors. The process can take up to 5 minutes for VM / 15 minutes for baremetal. If the process takes longer, see Hardware Introspection Problems.

Note

If you need to introspect just a single node, see Introspecting a Single Node

Provide Nodes

Only nodes in the available provisioning state can be deployed to (see Bare Metal Node States for details). To move nodes from manageable to available the following command can be used:

openstack overcloud node provide --all-manageable

Flavor Details

The undercloud will have a number of default flavors created at install time. In most cases these flavors do not need to be modified, but they can be if desired. By default, all overcloud instances will be booted with the baremetal flavor, so all baremetal nodes must have at least as much memory, disk, and cpu as that flavor.

In addition, there are profile-specific flavors created which can be used with the profile-matching feature. For more details on deploying with profiles, see Advanced Profile Matching.

Configure a nameserver for the Overcloud

Overcloud nodes can have a nameserver configured in order to resolve hostnames via DNS. The nameserver is defined in the undercloud’s neutron subnet. If needed, define the nameserver to be used for the environment:

# List the available subnets
openstack subnet list
openstack subnet set <subnet-uuid> --dns-nameserver <nameserver-ip>

Stable Branch

For Mitaka release and older, the subnet commands are executed within the neutron command:

neutron subnet-list
neutron subnet-update <subnet-uuid> --dns-nameserver <nameserver-ip>

Note

A public DNS server, such as 8.8.8.8 or the undercloud DNS name server can be used if there is no internal DNS server.

Virtual

In virtual environments, the libvirt default network DHCP server address, typically 192.168.122.1, can be used as the overcloud nameserver.

Deploy the Overcloud

Validations

Before you start the deployment, you may want to run the pre-deployment validations:

openstack tripleo validator run --group pre-deployment

Then verify the results as described in Running validation groups.

By default 1 compute and 1 control node will be deployed, with networking configured for the virtual environment. To customize this, see the output of:

openstack help overcloud deploy

Swap

Swap files or partitions can be installed as part of an Overcloud deployment. For adding swap files there is no restriction besides having 4GB available on / (by default). When using a swap partition, the partition must exist and be tagged as swap1 (by default). To deploy a swap file or partition in each Overcloud node use one of the following arguments when deploying:

-e /usr/share/openstack-tripleo-heat-templates/environments/enable-swap-partition.yaml
-e /usr/share/openstack-tripleo-heat-templates/environments/enable-swap.yaml

Ceph

When deploying Ceph with dedicated CephStorage nodes to host the CephOSD service it is necessary to specify the number of CephStorage nodes to be deployed and to provide some additional parameters to enable usage of Ceph for Glance, Cinder, Nova or all of them. To do so, use the following arguments when deploying:

--ceph-storage-scale <number of nodes> -e /usr/share/openstack-tripleo-heat-templates/environments/ceph-ansible/ceph-ansible.yaml

When deploying Ceph without dedicated CephStorage nodes, opting for an HCI architecture instead, where the CephOSD service is colocated with the NovaCompute service on the Compute nodes, use the following arguments:

-e /usr/share/openstack-tripleo-heat-templates/environments/hyperconverged-ceph.yaml -e /usr/share/openstack-tripleo-heat-templates/environments/ceph-ansible/ceph-ansible.yaml

The hyperconverged-ceph.yaml environment file will also enable a port on the StorageMgmt network for the Compute nodes. This will be the Ceph private network and the Compute NIC templates have to be configured to use that, see Configuring Network Isolation for more details on how to do it.

RHEL Satellite Registration

To register the Overcloud nodes to a Satellite add the following flags to the deploy command:

--rhel-reg --reg-method satellite --reg-org <ORG ID#> --reg-sat-url <satellite URL> --reg-activation-key <KEY>

Note

Only using an activation key is supported when registering to Satellite, username/password is not supported for security reasons. The activation key must enable the following repos:

rhel-7-server-rpms

rhel-7-server-optional-rpms

rhel-7-server-extras-rpms

rhel-7-server-openstack-6.0-rpms

SSL

To deploy an overcloud with SSL, see Deploying with SSL.

Run the deploy command, including any additional parameters as necessary:

openstack overcloud deploy --templates [additional parameters]

Note

When deploying a new stack or updating a preexisting deployment, it is important to avoid using component cli along side the unified cli. This will lead to unexpected results.

Example:

The following will present a behavior where the my_roles_data will persist, due to the location of the custom roles data, which is stored in swift:

openstack overcloud deploy --templates -r my_roles_data.yaml
heat stack-delete overcloud

Allow the stack to be deleted then continue:

openstack overcloud deploy --templates

The execution of the above will still reference my_roles_data as the unified command line client will perform a look up against the swift storage. The reason for the unexpected behavior is due to the heatclient lack of awareness of the swift storage.

The correct course of action should be as followed:

openstack overcloud deploy --templates -r my_roles_data.yaml
openstack overcloud delete <stack name>

Allow the stack to be deleted then continue:

openstack overcloud deploy --templates

To deploy an overcloud with multiple controllers and achieve HA, follow Configuring High Availability.

Virtual

When deploying the Compute node in a virtual machine without nested guest support, add --libvirt-type qemu or launching instances on the deployed overcloud will fail.

Note

To deploy the overcloud with network isolation, bonds, and/or custom network interface configurations, instead follow the workflow here to deploy: Configuring Network Isolation

Note

Previous versions of the client had many parameters defaulted. Some of these parameters are now pulling defaults directly from the Heat templates. In order to override these parameters, one should use an environment file to specify these overrides, via ‘parameter_defaults’.

The parameters that controlled these parameters will be deprecated in the client, and eventually removed in favor of using environment files.

Post-Deployment

Validations

After the deployment finishes, you can run the post-deployment validations:

openstack tripleo validator run --group post-deployment

Then verify the results as described in Running validation groups.

Deployment artifacts

Artifacts from the deployment, including log files, rendered templates, and generated environment files are saved under the working directory which can be specified with the --work-dir argument to openstack overcloud deploy. By default, the location is ~/overcloud-deploy/<stack>.

Access the Overcloud

openstack overcloud deploy generates an overcloudrc file appropriate for interacting with the deployed overcloud in the current user’s home directory. To use it, simply source the file:

source ~/overcloudrc

To return to working with the undercloud, source the stackrc file again:

source ~/stackrc

Add entries to /etc/hosts

In cases where the overcloud hostnames are not already resolvable with DNS, entries can be added to /etc/hosts to make them resolvable. This is particularly convenient on the undercloud. The Heat stack provides an output value that can be appended to /etc/hosts easily. Run the following command to get the output value and add it to /etc/hosts wherever the hostnames should be resolvable:

openstack stack output show overcloud HostsEntry -f value -c output_value

Setup the Overcloud network

Initial networks in Neutron in the overcloud need to be created for tenant instances. The following are example commands to create the initial networks. Edit the address ranges, or use the necessary neutron commands to match the environment appropriately. This assumes a dedicated interface or native VLAN:

openstack network create public --external --provider-network-type flat \
--provider-physical-network datacentre
openstack subnet create --allocation-pool start=172.16.23.140,end=172.16.23.240 \
--network public --gateway 172.16.23.251 --no-dhcp --subnet-range \
172.16.23.128/25 public

The example shows naming the network “public” because that will allow tempest tests to pass, based on the default floating pool name set in nova.conf. You can confirm that the network was created with:

openstack network list

Sample output of the command:

+--------------------------------------+----------+--------------------------------------+
| ID                                   | Name     | Subnets                              |
+--------------------------------------+----------+--------------------------------------+
| 4db8dd5d-fab5-4ea9-83e5-bdedbf3e9ee6 | public   | 7a315c5e-f8e2-495b-95e2-48af9442af01 |
+--------------------------------------+----------+--------------------------------------+

To use a VLAN, the following example should work. Customize the address ranges and VLAN id based on the environment:

openstack network create public --external --provider-network-type vlan \
--provider-physical-network datacentre --provider-segment 195
openstack subnet create --allocation-pool start=172.16.23.140,end=172.16.23.240 \
--network public --no-dhcp --gateway 172.16.23.251 \
--subnet-range 172.16.23.128/25 public

Validate the Overcloud

Check the Tempest documentation on how to run tempest.

Redeploy the Overcloud

The overcloud can be redeployed when desired.

  1. First, delete any existing Overcloud:

    openstack overcloud delete overcloud
    
  2. Confirm the Overcloud has deleted. It may take a few minutes to delete:

    # This command should show no stack once the Delete has completed
    openstack stack list
    
  3. It is recommended that you delete existing partitions from all nodes before redeploying, see Node cleaning for details.

  4. Deploy the Overcloud again:

    openstack overcloud deploy --templates