[RFE] Structured Tackle CR.

[jortel]

Restructure Tackle CRD for Better OpenShift Console Form Support

Problem

The current Tackle CRD uses x-kubernetes-preserve-unknown-fields: true with flat, snake_cased variables (e.g., feature_auth_required,
hub_container_limits_cpu, kai_llm_provider). This makes it difficult for the OpenShift console to generate a user-friendly form for configuring
Tackle instances.

Proposed Solution

Define a fully structured OpenAPI schema in the CRD with logical groupings that will enable the OpenShift console to display an organized, intuitive
form.

Proposed Structure

Below is an example Tackle CR using the proposed structured format:

apiVersion: tackle.konveyor.io/v1alpha1
kind: Tackle
metadata:
  name: tackle
  namespace: konveyor-tackle
spec:
  # Feature toggles
  features:
    authRequired: true
    authType: keycloak  # keycloak, rhsso, rhbk
    isolateNamespace: true
    analysisArchiver: true
    discovery: true

  # Hub component configuration
  hub:
    image: quay.io/konveyor/tackle2-hub:latest
    replicas: 1
    logLevel: 3
    metrics:
      enabled: true
      port: 2112
    resources:
      requests:
        cpu: 500m
        memory: 1Gi
      limits:
        cpu: "1"
        memory: 2Gi
    storage:
      database:
        size: 10Gi
        # storageClass: my-storage-class
      bucket:
        size: 100Gi
        accessMode: ReadWriteOnce
        # storageClass: my-storage-class
    tls:
      enabled: false
      # secretName: custom-tls-secret

  # UI component configuration
  ui:
    image: quay.io/konveyor/tackle2-ui:latest
    replicas: 1
    resources:
      requests:
        cpu: 100m
        memory: 350Mi
      limits:
        cpu: 500m
        memory: 800Mi
    ingress:
      enabled: true
      className: nginx
      proxyBodySize: 500m
      # pathType: Prefix
    route:
      enabled: true
      tlsTermination: edge
      insecureEdgeTerminationPolicy: Redirect
    tls:
      enabled: false

  # Authentication configuration
  auth:
    # Common auth settings
    adminUsername: admin

    # Keycloak/SSO settings (when features.authType is keycloak)
    keycloak:
      image: quay.io/keycloak/keycloak:latest
      replicas: 1
      realm: tackle
      clientId: tackle-ui
      apiAudience: konveyor-api
      requirePasswordUpdate: true
      javaOpts: "-Dcom.redhat.fips=false"
      resources:
        requests:
          cpu: 300m
          memory: 600Mi
        limits:
          cpu: "1"
          memory: 2Gi
      database:
        image: postgres:15
        name: keycloak_db
        version: "15"
        resources:
          requests:
            cpu: 100m
            memory: 350Mi
          limits:
            cpu: 500m
            memory: 800Mi
        storage:
          size: 1Gi
      tls:
        enabled: false
        # secretName: custom-tls-secret
      # hostname: keycloak.example.com
      # hostnameDynamic: false

    # RHSSO settings (when features.authType is rhsso)
    rhsso:
      externalAccess: false
      tls:
        enabled: true

    # RHBK settings (when features.authType is rhbk)
    rhbk:
      tls:
        enabled: true

  # Analysis components
  analyzer:
    image: quay.io/konveyor/addon-analyzer:latest
    resources:
      requests:
        cpu: "1"
        memory: 512Mi
      limits:
        cpu: "1"
        memory: 1Gi

    # Language providers
    providers:
      java:
        image: quay.io/konveyor/provider-java:latest
        resources:
          requests:
            cpu: "1"
            memory: 2.5Gi
          limits:
            cpu: "1"
            memory: 2.5Gi

      csharp:
        image: quay.io/konveyor/provider-csharp:latest
        resources:
          requests:
            cpu: "1"
            memory: 1Gi
          limits:
            cpu: "1"
            memory: 1Gi

      python:
        image: quay.io/konveyor/provider-generic:latest
        resources:
          requests:
            cpu: "1"
            memory: 1Gi
          limits:
            cpu: "1"
            memory: 1Gi

      nodejs:
        image: quay.io/konveyor/provider-generic:latest
        resources:
          requests:
            cpu: "1"
            memory: 1Gi
          limits:
            cpu: "1"
            memory: 1Gi

    # Discovery addon
    discovery:
      image: quay.io/konveyor/addon-discovery:latest
      resources:
        requests:
          cpu: 500m
          memory: 512Mi
        limits:
          cpu: 500m
          memory: 512Mi

    # Platform addon
    platform:
      image: quay.io/konveyor/addon-platform:latest
      resources:
        requests:
          cpu: "1"
          memory: 100Mi
        limits:
          cpu: "1"
          memory: 512Mi

  # Kai (AI/LLM) configuration
  kai:
    enabled: false
    image: quay.io/konveyor/kai:latest
    logLevel: info
    enableDemoMode: false
    enableTrace: true

    # LLM configuration
    llm:
      model: gpt-4
      provider: openai  # openai, azure, google, bedrock, anthropic
      baseUrl: https://api.openai.com/v1
      temperature: 0.7
      maxTokens: 2048
      maxRetries: 3
      # modelSpecificKwargs: {}

      # LLM Proxy (optional)
      proxy:
        enabled: false
        image: quay.io/lightspeed-core/lightspeed-stack:latest

    # Kai database
    database:
      image: postgres:15
      storage:
        size: 5Gi

  # Storage configuration
  storage:
    cache:
      enabled: true
      size: 100Gi
      accessMode: ReadWriteMany
      # storageClass: nfs

    # Global storage class override
    # storageClass: my-storage-class

  # Network configuration
  network:
    # HTTP/HTTPS proxy settings
    httpProxy: ""
    httpsProxy: ""
    noProxy: ""

    # OAuth proxy (for OpenShift)
    oauth:
      enabled: true
      provider: openshift
      image: quay.io/openshift/oauth-proxy:latest

  # Global settings
  imagePullPolicy: IfNotPresent

  # Advanced options
  options:
    # taskPodQuota: 50
    disableMavenSearch: false
    rwxSupported: false

Benefits

1. Better UX: OpenShift console can generate a multi-section form with clear categories
2. Validation: OpenAPI schema enables field validation, defaults, and documentation
3. Discoverability: Users can see all available options with descriptions
4. Type Safety: Proper typing (boolean, integer, string) prevents configuration errors
5. Maintainability: Logical grouping makes the CR easier to understand and maintain
6. Documentation: Schema descriptions appear as tooltips in the console

Implementation Plan

1. Update the CRD OpenAPI schema to define the full structure
2. Update the Ansible operator to support both old (flat) and new (structured) formats for backward compatibility
3. Add schema descriptions for each field to improve documentation
4. Add validation rules (min/max values, enums, required fields)
5. Update documentation and examples

Migration Strategy

To maintain backward compatibility:
- Support both old flat variables and new structured format
- Use a transition period where both are supported
- Provide migration guide and tools
- Default to structured format in new installations

Open Questions

1. Should we maintain backward compatibility indefinitely or deprecate flat variables?
2. What's the preferred migration timeline?
3. Should some rarely-used advanced options remain flat or be nested under an advanced section?

This structure groups related configuration together, making it much easier to work with in the OpenShift console while maintaining clarity about what
 each component does.

[konveyor-ci-bot]

This issue is currently awaiting triage.
If contributors determine this is a relevant issue, they will accept it by applying the triage/accepted label and provide further guidance.
The triage/accepted label can be added by org members.

[shawn-hurley]

I think this is a really good idea. I do wonder if we want to consider an enhancement for this?

I don't know if we do, because the impact should be limited to this repo but was just a thought.

[github-actions]

This issue has been automatically marked as stale because it has not had any activity for 60 days.
It will remain open for visibility and reporting purposes.
Please comment if this issue is still relevant.