MAS DevOps Ansible Collection Ansible CLI
Home Roles: Utilities gencfg_watsonstudio Role Variables Example Playbook License

gencfg_watsonstudio¤

Generate Watson Studio configuration files for connecting Maximo Application Suite to IBM Cloud Pak for Data Watson Studio instances. This role creates WatsonStudioCfg custom resources that enable MAS applications (particularly Predict and Health) to leverage Watson Studio's machine learning and analytics capabilities.

The role supports flexible configuration scoping to make Watson Studio available at system, workspace, application, or workspace-application levels within MAS.

Role Variables¤

mas_instance_id¤

MAS instance identifier for which the Watson Studio configuration is being generated.

  • Required
  • Environment Variable: MAS_INSTANCE_ID
  • Default: None

Purpose: Identifies the target MAS instance where this Watson Studio configuration will be applied.

When to use: Always required when generating Watson Studio configuration. Must match an existing MAS instance ID.

Valid values: Valid MAS instance ID (typically 3-12 lowercase alphanumeric characters).

Impact: The generated configuration will be namespace-scoped to mas-<instance-id>-core. Incorrect instance ID will cause the configuration to be created in the wrong namespace.

Related variables: mas_config_dir, mas_config_scope

Notes: - Must match the instance ID used during MAS installation - Case-sensitive value - Watson Studio integration is primarily used by Predict and Health applications - Required for AI/ML model training and deployment in MAS

mas_workspace_id¤

MAS workspace identifier for workspace-scoped configurations.

  • Required if mas_config_scope is ws or wsapp
  • Environment Variable: MAS_WORKSPACE_ID
  • Default: None

Purpose: Specifies the target workspace when generating workspace-scoped or workspace-application-scoped Watson Studio configurations.

When to use: Required when mas_config_scope is set to ws or wsapp. Not used for system or app scopes.

Valid values: Valid MAS workspace ID (typically lowercase alphanumeric, e.g., masdev, prod, test).

Impact: The Watson Studio configuration will only be available to the specified workspace. Applications in other workspaces cannot access this Watson Studio instance.

Related variables: mas_config_scope, mas_application_id

Notes: - Workspace must exist before applying the configuration - Use workspace-scoped configs for tenant isolation in multi-tenant deployments - Each workspace can have its own Watson Studio instance for data isolation

mas_config_scope¤

Configuration scope level for the generated Watson Studio configuration.

  • Optional
  • Environment Variable: MAS_CONFIG_SCOPE
  • Default: system

Purpose: Determines at what level the Watson Studio configuration will be available within MAS - system-wide, workspace-specific, application-specific, or workspace-application combination.

When to use: Set based on how Watson Studio will be shared across MAS components. Default system scope makes it available to all workspaces and applications.

Valid values: - system - Available to all workspaces and applications (default) - ws - Available only to a specific workspace - app - Available only to a specific application across all workspaces - wsapp - Available only to a specific application in a specific workspace

Impact: Determines which MAS components can access this Watson Studio configuration. Incorrect scope may prevent applications from finding the Watson Studio connection.

Related variables: mas_workspace_id (required for ws and wsapp), mas_application_id (required for app and wsapp)

Notes: - Default system scope is suitable for most deployments - Use wsapp scope for dedicated Watson Studio instances per workspace-application - Scope cannot be changed after creation; requires recreation of the configuration - Predict and Health are the primary consumers of Watson Studio

mas_config_dir¤

Local directory path where the generated YAML configuration file will be saved.

  • Required
  • Environment Variable: MAS_CONFIG_DIR
  • Default: None

Purpose: Specifies the output directory for the generated WatsonStudioCfg YAML file that can be applied to the MAS instance.

When to use: Always required. Directory must exist or be creatable by the Ansible user.

Valid values: Valid directory path (absolute or relative) where the Ansible controller has write permissions.

Impact: The generated YAML file will be created in this directory and can be applied using kubectl apply or the suite_config role.

Related variables: mas_instance_id

Notes: - Directory will be created if it doesn't exist - Generated filename format: watsonstudiocfg-<scope>.yml - Keep generated files for documentation and disaster recovery purposes - Can be used as input to the suite_config role

mas_application_id¤

MAS application identifier for application-scoped configurations.

  • Required if mas_config_scope is app or wsapp
  • Environment Variable: MAS_APP_ID
  • Default: None

Purpose: Specifies the target application when generating application-scoped or workspace-application-scoped Watson Studio configurations.

When to use: Required when mas_config_scope is set to app or wsapp. Not used for system or ws scopes.

Valid values: Valid MAS application ID, typically predict or health (the primary Watson Studio consumers).

Impact: The Watson Studio configuration will only be available to the specified application. Other applications cannot access this Watson Studio instance.

Related variables: mas_config_scope, mas_workspace_id

Notes: - Application must be installed before applying the configuration - Predict and Health are the main applications that use Watson Studio - Use application-scoped configs when different applications need different Watson Studio instances

cpd_admin_username¤

Cloud Pak for Data administrative username for Watson Studio access.

  • Required
  • Environment Variable: CPD_ADMIN_USERNAME
  • Default: None

Purpose: Specifies the Cloud Pak for Data user account that MAS will use to connect to Watson Studio for AI/ML operations.

When to use: Always required. Must be a valid CP4D user with appropriate Watson Studio permissions.

Valid values: Valid Cloud Pak for Data username with Watson Studio access rights.

Impact: This user must have permissions to create and manage Watson Studio projects, deployments, and models. Insufficient permissions will cause Predict/Health AI features to fail.

Related variables: cpd_admin_password, cpd_admin_url

Notes: - User must have Watson Studio service access - Requires permissions to create projects and deploy models - Consider using a dedicated service account rather than a personal admin account - User should have appropriate resource quotas for ML workloads

cpd_admin_password¤

Cloud Pak for Data password for authentication.

  • Required
  • Environment Variable: CPD_ADMIN_PASSWORD
  • Default: None

Purpose: Provides the password for the Cloud Pak for Data user account specified in cpd_admin_username.

When to use: Always required for CP4D authentication.

Valid values: Valid password string meeting Cloud Pak for Data security requirements.

Impact: Stored as a Kubernetes secret in the MAS namespace. Incorrect password will prevent MAS from connecting to Watson Studio.

Related variables: cpd_admin_username, cpd_admin_url

Notes: - Password is stored securely in Kubernetes secrets - Use strong passwords meeting your organization's security policies - Consider implementing password rotation policies - Ensure password doesn't expire or require periodic changes that could break integration

cpd_admin_url¤

Cloud Pak for Data console URL for Watson Studio access.

  • Required
  • Environment Variable: CPD_ADMIN_URL
  • Default: None

Purpose: Defines the Cloud Pak for Data web console URL that MAS will use to connect to Watson Studio services.

When to use: Always required. Must be the accessible URL to the CP4D console.

Valid values: Valid HTTPS URL to the Cloud Pak for Data console (e.g., https://cpd-cpd.apps.ocp.example.com).

Impact: MAS applications will use this URL to access Watson Studio APIs. Incorrect or inaccessible URL will prevent AI/ML functionality.

Related variables: cpd_admin_username, cpd_admin_password

Notes: - Must be the full HTTPS URL including protocol - URL must be accessible from the OpenShift cluster where MAS is running - Verify network connectivity and firewall rules - For multi-cluster deployments, ensure proper routing is configured - URL should point to the CP4D console route, not individual service routes

custom_labels¤

Custom Kubernetes labels to apply to the generated WatsonStudioCfg resource.

  • Optional
  • Environment Variable: CUSTOM_LABELS
  • Default: None

Purpose: Adds custom labels to the WatsonStudioCfg resource for organization, filtering, or automation purposes.

When to use: Use when you need to tag configurations for specific purposes like environment identification, cost tracking, or automated management.

Valid values: Comma-separated key=value pairs (e.g., env=prod,team=ai-ml,cost-center=12345).

Impact: Labels are applied to the Kubernetes resource and can be used for selection, filtering, and automation workflows.

Related variables: None

Notes: - Labels must follow Kubernetes naming conventions - Useful for GitOps workflows and resource management - Can be used with label selectors in automation scripts - Helpful for tracking Watson Studio configurations across multiple environments

Example Playbook¤


- hosts: localhost
  any_errors_fatal: true
  roles:
    - ibm.mas_devops.gencfg_watsonstudio

License¤

EPL-2.0