Installation¶

Local cluster¶

There are a number of tools that enable you to create a local Kubernetes cluster. In this guide, the cluster will be created with kind.

To create a new cluster with kind:

1. First, create a single-node cluster. Note that since the node is running inside a Docker container, the extraPortMappings section is required to map the container ports to the host in order to able to access the Ingress controller.

   cat <<EOF | kind create cluster --config=-
   kind: Cluster
   apiVersion: kind.x-k8s.io/v1alpha4
   nodes:
   - role: control-plane
     kubeadmConfigPatches:
     - |
       kind: InitConfiguration
       nodeRegistration:
         kubeletExtraArgs:
           node-labels: "ingress-ready=true"
     extraPortMappings:
     - containerPort: 80
       hostPort: 80
       protocol: TCP
     - containerPort: 443
       hostPort: 443
       protocol: TCP
   EOF
   

2. Deploy the Nginx Ingress controller.

   kubectl apply --filename https://raw.githubusercontent.com/kubernetes/ingress-nginx/main/deploy/static/provider/kind/deploy.yaml
   

Licenses¶

Ontotext Metadata Studio and GraphDB Enterprise Edition require a license to function. These should be provided as secrets in the cluster before installing.

To create the secrets, execute the following commands:

kubectl create secret generic metadata-studio-license \
    --from-file metadata-studio.license=/path/to/metadata-studio.license

kubectl create secret generic graphdb-license \
    --from-file graphdb.license=/path/to/graphdb-ee.license

Changing the Ingress host¶

By default, the address on the application will be http://omds.127.0.0.1.nip.io. This is suitable for local previews, but not for production.

To deploy the application on a different domain, certain configuration properties need to be changed:

1. Create a file called values.overrides.yaml.

2. Considering that Metadata Studio needs to be accessible at http://metadata-studio.example.com, place the following content in the file:

   ingress:
     host: metadata-studio.example.com
   
   graphdb:
     deployment:
       host: metadata-studio.example.com
   
   web:
     configuration:
       OMDS_OAUTH_REDIRECT_URI: http://metadata-studio.example.com/oauth
       OMDS_AUTHORIZATION_ENDPOINT: http://metadata-studio.example.com/auth/realms/metadata-studio/protocol/openid-connect/auth
       OMDS_TOKEN_ENDPOINT: http://metadata-studio.example.com/auth/realms/metadata-studio/protocol/openid-connect/token
       OMDS_END_SESSION_ENDPOINT: http://metadata-studio.example.com/auth/realms/metadata-studio/protocol/openid-connect/logout
   
   keycloak:
     realm:
       rootUrl: http://metadata-studio.example.com
   

3. Now, Ontotext Metadata Studio can be installed with the following command:

   helm upgrade --install --timeout 15m --values /path/to/values.overrides.yaml metadata-studio ontotext/metadata-studio
   

Configuring Tagging service API key¶

The Ontotext Tagging service requires an API to function. To provide the key, include it in the values.overrides.yaml file:

service:
  configurationSecret:
    tag.api.key: my-tag-api-key

Changing the default Metadata Studio logo¶

Metadata Studio provides the option to change its default logo.

To deploy the application with a different logo:

1. Create a configmap with the following structure (the below K8S configmap name serves as an example):

   apiVersion: v1
   kind: ConfigMap
   metadata:
     name: omds-logo-config
   binaryData:
     logo.png:
       {{ .Files.Get "files/logo.png" | b64enc }}
     favicon.ico:
       {{ .Files.Get "files/favicon.ico" | b64enc }}
   

2. Provide the files mentioned in the configmap.

3. Create a file called values.overrides.yaml.

4. In it, place the following content:

   logo:
     override: true
     configmap: omds-logo-config
   

5. Now, Metadata Studio can be installed with the following command:

   helm upgrade --install --timeout 15m --values /path/to/values.overrides.yaml metadata-studio ontotext/metadata-studio