AI Service Installation¶

Usage¶

For full usage information run mas aiservice-install --help

The mas aiservice-install command is specifically designed for standalone installation of AI Service version 9.1.x or above. This command provides a streamlined installation process focused solely on AI Service and its dependencies.

Preparation¶

IBM Entitlement Key¶

Access Container Software Library using your IBMId to obtain your entitlement key.

MAS License File¶

Access IBM License Key Center, on the Get Keys menu select IBM AppPoint Suites. Select IBM MAXIMO APPLICATION SUITE AppPOINT LIC and on the next page fill in the information as below:

Field	Content
Number of Keys	How many AppPoints to assign to the license file
Host ID Type	Set to Ethernet Address
Host ID	Enter any 12 digit hexadecimal string
Hostname	Set to the hostname of your OCP instance, but this can be any value really.
Port	Set to 27000

The other values can be left at their defaults. Finally, click Generate and download the license file to your home directory as entitlement.lic.

OpenShift Cluster¶

You should already have a target OpenShift cluster ready to install AI Service into. If you do not already have one then refer to the OpenShift Container Platform installation overview.

The CLI also supports OpenShift provisioning in many hyperscaler providers:

• AWS
• IBM Cloud
• IBM DevIT FYRE (Internal)

Operator Catalog Selection¶

If you have not already determined the catalog version for your installation, refer to the information in the Operator Catalog topic, or contact IBM Support for guidance.

Interactive Install¶

Run the mas aiservice-install command and follow the interactive prompts. Mount your home directory to access the license file when needed.

docker run -ti --rm -v ~:/mnt/home quay.io/ibmmas/cli:18.10.0 mas aiservice-install

The interactive install will guide you through the following steps:

Non-Interactive Install¶

The following command demonstrates a complete AI Service installation with all configuration options. This command will launch the MAS CLI container image, login to your OpenShift Cluster and start the AI Service installation without triggering any prompts.

IBM_ENTITLEMENT_KEY=your_entitlement_key_here

docker run -e IBM_ENTITLEMENT_KEY -ti --rm -v ~:/mnt/home quay.io/ibmmas/cli:18.10.0 bash -c "
  oc login --token=sha256~xxxx --server=https://xxx &&
  mas aiservice-install \
    --mas-catalog-version v9-260129-amd64 \
    --ibm-entitlement-key \${IBM_ENTITLEMENT_KEY} \
    --aiservice-instance-id aiservice1 \
    --aiservice-channel 9.1.x \
    \
    --storage-class-rwo nfs-client \
    --storage-class-rwx nfs-client \
    --storage-pipeline nfs-client \
    --storage-accessmode ReadWriteMany \
    \
    --license-file /mnt/home/entitlement.lic \
    --uds-email admin@example.com \
    --uds-firstname John \
    --uds-lastname Doe \
    \
    --dro-namespace redhat-marketplace \
    --mongodb-namespace mongoce \
    \
    --s3-tenants-bucket km-tenants \
    --s3-templates-bucket km-templates \
    \
    --odh-model-deployment-type serverless \
    \
    --watsonxai-apikey your_watsonx_api_key \
    --watsonxai-url https://us-south.ml.cloud.ibm.com \
    --watsonxai-project-id your_project_id \
    \
    --install-minio
    --minio-root-user minio \
    --minio-root-password minio123 \
    \
    --tenant-entitlement-type standard \
    --tenant-entitlement-start-date 2025-01-01 \
    --tenant-entitlement-end-date 2026-01-01 \
    \
    --rsl-url http://your-rsl-host:3001/api/v3/vector/query \
    --rsl-org-id your_org_id \
    --rsl-token 'Bearer your_rsl_token' \
    \
    --accept-license --no-confirm
"

Core Parameters¶

Parameter	Description	Required	Example
--mas-catalog-version	MAS operator catalog version	Yes	v9-250902-amd64
--aiservice-instance-id	Unique identifier for AI Service instance	Yes	aiservice1
--aiservice-channel	AI Service version channel	Yes	9.1.x
--ibm-entitlement-key	IBM Container Registry entitlement key	Yes	$IBM_ENTITLEMENT_KEY

Storage Configuration¶

Parameter	Description	Required	Example
--storage-class-rwo	ReadWriteOnce storage class	Yes	nfs-client
--storage-class-rwx	ReadWriteMany storage class	Yes	nfs-client
--storage-pipeline	Storage class for pipeline workloads	Yes	nfs-client
--storage-accessmode	Storage access mode	Yes	ReadWriteMany

License and Contact Information¶

Parameter	Description	Required	Example
--license-file	Path to MAS license file (required if SLS not installed)	Conditional	/mnt/home/entitlement.lic
--uds-email	Contact email address	Yes	admin@example.com
--uds-firstname	Contact first name	Yes	John
--uds-lastname	Contact last name	Yes	Doe

Dependency Configuration¶

Parameter	Description	Required	Example
--dro-namespace	IBM Data Reporter Operator namespace	Optional	redhat-marketplace
--mongodb-namespace	MongoDB namespace (for SLS)	Optional	mongoce

S3 Storage Configuration¶

Parameter	Description	Required	Example
--s3-accesskey	S3 access key	Yes	minio
--s3-secretkey	S3 secret key	Yes	minio123
--s3-host	S3 service host	Yes	minio-service.minio.svc.cluster.local
--s3-port	S3 service port	Yes	9000
--s3-ssl	Enable SSL for S3 connection	Yes	false
--s3-region	S3 region	Yes	none
--s3-bucket-prefix	Prefix for S3 buckets	Yes	s3-
--s3-tenants-bucket	S3 bucket for tenant data	Yes	km-tenants
--s3-templates-bucket	S3 bucket for templates	Yes	km-templates

OpenDataHub Configuration¶

Parameter	Description	Required	Example
--odh-model-deployment-type	Model deployment type (serverless/raw)	Yes	serverless

Watson AI Configuration¶

Parameter	Description	Required	Example
--watsonxai-apikey	Watson AI API key	Optional	your_api_key
--watsonxai-url	Watson AI service URL	Optional	https://us-south.ml.cloud.ibm.com
--watsonxai-project-id	Watson AI project ID	Optional	your_project_id

MinIO Configuration¶

Parameter	Description	Required	Example
--minio-root-user	MinIO root username	Yes	minio
--minio-root-password	MinIO root password	Yes	minio123

Tenant Configuration¶

Parameter	Description	Required	Example
--tenant-entitlement-type	Tenant entitlement type	Yes	standard
--tenant-entitlement-start-date	Entitlement start date (YYYY-MM-DD)	Yes	2025-01-01
--tenant-entitlement-end-date	Entitlement end date (YYYY-MM-DD)	Yes	2026-01-01

RSL (Red Hat Service Locator) Configuration¶

Parameter	Description	Required	Example
--rsl-url	RSL service URL	Optional	http://host:3001/api/v3/vector/query
--rsl-org-id	RSL organization ID	Optional	your_org_id
--rsl-token	RSL authentication token	Optional	Bearer your_token

Additional Options¶

Parameter	Description	Required
--accept-license	Accept license terms	Yes
--no-confirm	Skip confirmation prompts	Optional

Dependencies¶

The AI Service installation automatically handles the following dependencies:

Required Dependencies¶

• IBM Suite License Service (SLS): Manages licensing for AI Service. Automatically installed if not present, or can use existing SLS instance.
• IBM Data Reporter Operator (DRO): Handles data reporting and metrics
• Red Hat Certificate Manager: Manages TLS certificates
• MinIO: Provides S3-compatible object storage for AI Service data
• Db2: Database for AI Service metadata and configuration

Optional Dependencies¶

• MongoDB Community Edition: Only required if installing SLS for the first time

More Information¶

The AI Service installation is designed to work on any OCP cluster, but has been specifically tested in these environments:

• IBMCloud ROKS
• Microsoft Azure
• AWS ROSA
• IBM DevIT FYRE (internal)

The engine that performs all tasks is written in Ansible. The code is open source and available in ibm-mas/ansible-devops, and the collection is also available to install directly from Ansible Galaxy.

The installation is performed inside your RedHat OpenShift cluster utilizing OpenShift Pipelines:

OpenShift Pipelines is a Kubernetes-native CI/CD solution based on Tekton. It builds on Tekton to provide a CI/CD experience through tight integration with OpenShift and Red Hat developer tools. OpenShift Pipelines is designed to run each step of the CI/CD pipeline in its own container, allowing each step to scale independently to meet the demands of the pipeline.

Integration with Maximo Manage¶

After installing AI Service, you can bind it to Maximo Manage to enable AI capabilities:

During Manage Installation:

When installing Manage alongside AI Service using mas install:

• Installing AI Service with Manage (using --aiservice-channel):
• The binding is configured automatically with default "user" tenant
• No additional parameters needed

• Any --manage-aiservice-instance-id or --manage-aiservice-tenant-id parameters will be ignored

• Using Existing AI Service (not using --aiservice-channel):

• Interactive Mode: You will be prompted to select the AI Service instance and tenant
• Non-Interactive Mode: Use --manage-aiservice-instance-id and --manage-aiservice-tenant-id parameters

Post-Installation:

You can configure the AI Service binding through the Maximo Manage UI.

For more information on integrating AI Service with Manage, see the Installation Guide.