Skip to main content

Add an application to the ML platform

Last update:

In the ML platform, you can add additional applications using kubectl, helm, kustomize.

You can open the app via a URL like https://myapp-yourdomain.mlops.selcloud.ru or add the app to the ML platform start page.

For applications that will be available online, you need to set up authorization.

Add a new application to the ML platform⁠

To add a new application to the ML platform, you need to create an Ingress object. You do not need to create an Ingress Controller — a Traefik controller is pre-installed in the ML-platform in Managed Kubernetes clusters.

It is not necessary to obtain TLS certificates directly in the ML platform because the certificate is installed on the reverse proxy.

  1. Connect to the Managed Kubernetes cluster that was created when ML platform.

  2. Create a yaml file with a manifest for the Ingress object.

    Manifesto example:

    apiVersion: networking.k8s.io/v1
    kind: Ingress
    metadata:
      name: myapp
      namespace: <ml_platform_namespace>
      annotations:
        traefik.ingress.kubernetes.io/router.tls: "true"
    spec:
      tls:
        - hosts:
            - "myapp-<ml_platform_domain>"
          secretName: myapp-<ml_platform_domain>-cert
      rules:
        - host: "myapp-<ml_platform_domain>"
          http:
            paths:
              - path: /
                pathType: Prefix
                backend:
                  service:
                    name: myapp
                    port:
                      number: 80

    Specify:

    • <ml_platform_namespace> — Namespace of the ML platform;
    • <ml_platform_domain> — URL of the form yourdomain.mlops.selcloud.ru which was given after connecting ML-platform.

  3. Create Ingress:

    kubectl apply -f <ingress.yaml>

    Specify <ingress.yaml> is the name of the yaml file with the manifest for Ingress.

  4. Open the application at:

    https://myapp-<ml_platform_domain>

    Specify <ml_platform_domain> — URL of the form yourdomain.mlops.selcloud.ru, which was issued after connecting ML-platform.

  5. Configure authorization for the application.

  6. Optional: add the app to your start page.

Add an application to the ML-platform start page⁠

The ML platform start page is powered by the Forecastle tool. On the page you can see all the applications that are running in the Managed Kubernetes cluster by default.

If you've added a new app to the ML platform, it can also be placed on the home page.

  1. Connect to the Managed Kubernetes cluster that was created when ML platform.

  2. Open the yaml file with the manifest for the Ingress application and add annotations to it:

    apiVersion: networking.k8s.io/v1
    kind: Ingress
    metadata:
      name: myapp
      namespace: <ml_platform_namespace>
      annotations:
        traefik.ingress.kubernetes.io/router.tls: "true"
        forecastle.stakater.com/expose: "true"
        forecastle.stakater.com/appName: MyApp # Название приложения, которое будет показано на стартовой странице
        forecastle.stakater.com/group: MyAppsGroup # Группа на стартовой странице, в которую добавится приложение
        forecastle.stakater.com/icon: <app_icon_url>

    Specify:

    • <ml_platform_namespace> — Namespace of the ML platform;
    • <app_icon_url> — optional: URL of the image for the app icon.

  3. Apply the changes for Ingress:

    kubectl apply -f <ingress.yaml>

    Specify <ingress.yaml> is the name of the yaml file with the manifest for Ingress.

  4. Open the ML platform start page and check that the app has been added:

    https://myapp-<ml_platform_domain>

    Specify <ml_platform_domain> — URL of the form yourdomain.mlops.selcloud.ru, which was issued after connecting ML-platform.

Configure authorization for the application⁠

If you've added an application to the ML platform, be sure to set up authorization.

The authorization setting depends on the protocols that the application supports:

  • If the application supports authorization using OIDC/OAuth2/SAML protocols, create a Keycloak client;
  • If the application does not support OIDC/OAuth2/SAML protocols or the application does not have authorization mechanisms, use gogatekeeper — it is a sidecar for Keycloak.

Create a Keycloak client and configure the application to authorize through the Keycloak ML platform. When the user authorizes in the application, a request will be sent to the Keycloak client. If it fails it, the user will be able to log into the app. For more information about using Keycloak in ML-platform, see the Managing Users in Keycloak instructions.

  1. Connect to the Managed Kubernetes cluster that was created when ML platform.

  2. Create a Keycloak client through the Keycloak control panel at https://keycloak-<ml_platform_domain>/admin/cmlp/console/ or create a yaml file with a manifest for the KeycloakClient object.

    Manifesto example:

    apiVersion: keycloak.org/v1alpha1
    kind: KeycloakClient
    metadata:
      name: myapp-client
      namespace: <ml_platform_namespace>
    spec:
      client:
        # Settings
        enabled: true
        clientId: "<appclient_name>"
        name: ""
        description: ''
        secret: "<password>"
        protocol: "openid-connect"
        redirectUris:
          - "https://<appclient_name>-.<ml_platform_namespace>/*" # URI, с которого будет происходить редирект в Keycloak
        rootUrl: "${authBaseUrl}"
        baseUrl: "/"
        publicClient: false
        bearerOnly: false
        serviceAccountsEnabled: false
        consentRequired: false
        directAccessGrantsEnabled: true
        implicitFlowEnabled: false
        frontchannelLogout: false
        standardFlowEnabled: true
        surrogateAuthRequired: false
        useTemplateConfig: true
        useTemplateMappers: true
        # Scopes
        useTemplateScope: true
        fullScopeAllowed: false
        defaultClientScopes:
          - "profile"
          - "email"
        # Roles
        protocolMappers:
        - config:
            access.token.claim: "true"
            id.token.claim: "false"
            included.custom.audience: <appclient_name>
          consentRequired: false
          name: Audience-forecastle-cmlp
          protocol: openid-connect
          protocolMapper: oidc-audience-mapper
      realmSelector:
        matchLabels:
          app.kubernetes.io/name: keycloak-realm-cmlp
          app.kubernetes.io/instance: keycloak-operator

    Specify:

    • <ml_platform_namespace> — Namespace of the ML platform;
    • <appclient_name> — the unique name of the Keycloak client;
    • <password> — password for the Keycloak client. Required to configure the application.

  3. Create a Keycloak client:

    kubectl apply -f <keycloakclient.yaml>

    Specify <keycloakclient.yaml> is the name of the manifest yaml file to create the Keycloak client.

  4. Verify that the client has been created. Open the Security Admin Console application and go to Configure → Clients.

  5. Configure the application to authorize through the created Keycloak client:

    • use the python library python-keycloak;
    • or use Grafana and modify the configuration file following the example in the Grafana documentation.