5 Alternative deployments #

On your management workstation, clone the SUSE AI Node Installer repository.

> git clone https://github.com/SUSE/suse-ai-node-ansible.git
> cd suse-ai-node-ansible

Build the Docker image from the source.

> docker build \
  -t suse-ai-node-ansible-runner
  -f Dockerfile.local .

Or, if you prefer the newer buildx:

> docker buildx build \
  -t suse-ai-node-ansible-runner
  -f Dockerfile.local --load .

For each RKE2 cluster, create the inventory.ini file.

> cp inventory.ini.example inventory.ini

For each RKE2 cluster, create the extra_vars.yml file and align its entries with your use case. Refer to Section 5.1.3.1, “extra_vars.yml examples tailored to specific use cases” for examples of extra_vars.yml tailored to a specific use case.

> cp extra_vars.yml.example extra_vars.yml

Run the site.yml playbook. At a high level, the playbook automates the node setup by performing the following steps:

> docker run --rm \
  -v ~/.ssh/id_rsa:/root/.ssh/id_rsa:ro \
  -v ./inventory.ini:/workspace/inventory.ini \
  -v ./extra_vars.yml:/workspace/extra_vars.yml \
  suse-ai-node-ansible-runner \
  ansible-playbook -i inventory.ini playbooks/site.yml -e "@extra_vars.yml"

If your target ansible_host is localhost:

> docker run --rm \
  --network host \
  -v ~/.ssh/id_rsa:/root/.ssh/id_rsa:ro \
  -v ./inventory.ini:/workspace/inventory.ini \
  -v ./extra_vars.yml:/workspace/extra_vars.yml \
  suse-ai-node-ansible-runner \
  ansible-playbook -i inventory.ini playbooks/stage1.yml -e "@extra_vars.yml"

> docker run --rm \
  -v ~/.ssh/id_rsa:/root/.ssh/id_rsa:ro \
  -v ./inventory.ini:/workspace/inventory.ini \
  -v ./extra_vars.yml:/workspace/extra_vars.yml \
  suse-ai-node-ansible-runner \
  ansible-playbook -i inventory.ini playbooks/stage2.yml -e "@extra_vars.yml"

Note

NVIDIA drivers are not installed when localhost is the target. For localhost deployments, we recommend installing the drivers manually.

In Rancher, navigate to Extensions › Manage Extensions Catalog

Import the extension catalog from GitHub. Enter ghcr.io/suse/suse-ai-lifecycle-manager:<VERSION> in the Catalog Image Reference input field. Replace <VERSION> with a tag published in the GitHub Container Registry. Confirm with Load.

From the Manage Repositories page, verify if the SUSE AI Lifecycle Manager repository has the Active state. If not, refresh the connection.

Go back to the Extensions page and start the installation by clicking SUSE AI Lifecycle Manager.

Install the SUSE AI Operator:

> helm install suse-ai-operator \
  -n suse-ai-operator-system \ 1
  --version 0.1.0 \ 2
  --create-namespace \
  oci://ghcr.io/suse/chart/suse-ai-operator

Create the InstallAIExtension Custom Resource (CR). Save the suse-ai-extension.yaml file with the following content.

apiVersion: ai-platform.suse.com/v1alpha1
kind: InstallAIExtension
metadata:
  name: suseai
spec:
  helm:
    name: suse-ai-lifecycle-manager
    url: "oci://ghcr.io/suse/chart/suse-ai-lifecycle-manager"
    version: "1.0.0" 1
  extension:
    name: suse-ai-lifecycle-manager
    version: "1.0.0" 2

Apply the suse-ai-extension.yaml file to install the extension.

> kubectl apply -f suse-ai-extension.yaml

Refer to the document Rancher Authentication to create the user name and access token (password) for SUSE Application Collection.

In Rancher, navigate to Cluster › Apps › Repositories.

Click Create and provide the following details:

Wait until the repository status changes to Active.

Refer to the document Using container images from the SUSE Registry to get the user name and the access token (password) for SUSE Registry. The access code is the same as the registration code of your SUSE AI subscription, which you can find on your SCC account at https://scc.suse.com/ by following these steps:

In Rancher, navigate to Cluster › Apps › Repositories.

Click Create and provide the following details:

Wait until the repository status changes to Active.

Click the application’s tile to navigate to its instances page.

Provide the basic information required for the Helm installation, such as the release name, namespace and version. The repository is automatically selected. Click Next to navigate to the Target Cluster page to select the cluster(s).

On the Target Cluster page, choose one or more clusters that meet the application’s resource requirements and click Next to go to the Configuration page.

On the Configuration page, customize the installation using the UI form for key settings or the YAML file for advanced options, and click Next to go to the Review page.

Review the installation summary on the Review page and confirm by clicking Install to install the application. To navigate to previous steps to adjust the settings instead, click the Previous button.

Click the application’s tile to navigate to its instances page.

Click Manage for the application to start its Manage wizard.

Update the application’s details on the wizard screens.

On the Review page, click Save to apply the changes.

Click the application’s tile to navigate to its instances page.

Click Delete for the instance you want to delete.

In the pop-up window, confirm the deletion.