Configuration Reference

This document provides a complete reference for configuring the TSG CLI tool, including all available options for both ecosystem and participant deployments.

Configuration Files

The TSG CLI uses YAML configuration files to define deployment parameters:

• ecosystem.yaml - For deploying complete dataspace ecosystems with multiple participants
• participant.yaml - For deploying single participants to join existing dataspaces

Schema Overview

Both configuration types follow a specific schema validated by the CLI. The configuration is based on TypeScript models that ensure proper validation.

Configuration

Configuration for the CLI tool starts either at Ecosystem or SingleParticipant.

Ecosystem (ecosystem.yaml)

Name	Data Type	Required	Explanation	Default
general	General	Yes	General configuration properties
applications	Applications		Application configuration properties
participants	Array<Participant>	Yes	Participant configuration

SingleParticipant (participant.yaml)

Name	Data Type	Required	Explanation	Default
general	General	Yes	General configuration properties
applications	Applications		Application configuration properties
participant	Participant	Yes	Participant configuration

General

Name	Data Type	Required	Explanation	Default
namespace	String	Yes	Kubernetes namespace used for deployments. Must match pattern [a-zA-Z-]+ (alphanumeric with hyphens)
username	String	Yes	Default admin username
password	String	Yes	Default admin password
authorityDomain	String	Yes	Domain name of the authority, either the one deployed as participant or an external dataspace authority
credentialType	String	Yes	Credential type name
postgresDeploymentMode	"per-participant" | "per-namespace"		PostgreSQL deployment strategy. per-participant deploys a database per participant, per-namespace deploys a single shared cluster for the namespace	"per-namespace"
postgresInstances	Number		Number of PostgreSQL instances for shared cluster (per-namespace mode only). Higher values provide better high availability	1
postgresStorageSize	String		Storage size for shared PostgreSQL cluster (per-namespace mode only). Examples: "1Gi", "5Gi", "10Gi"	"1Gi"

Applications

Name	Data Type	Required	Explanation	Default
postgres	Application		PostgreSQL application using CloudNativePG
ssoBridge	Application		SSO Bridge application
wallet	Application		TSG Wallet
controlPlane	Application		TSG Control Plane
dataPlanes	Map<String, Application>		Data plane applications

Application

Name	Data Type	Required	Explanation	Default
chartVersion	String		Helm chart version
chartName	String		Helm chart name
developmentChart	Boolean		Use TSG development Helm repository	false
imageTag	String		Docker image tag
imageRepository	String		Docker image repository

Participant

Name	Data Type	Required	Explanation	Default
host	String	Yes	Domain name, used for host ingress paths
id	String	Yes	Participant identifier, used for identifier for instances
name	String	Yes	Participant name, used for metadata
routing	"path" | "subdomain"		Path-based or subdomain-based routing for services. Must be either "path" or "subdomain"	"path"
hasControlPlane	Boolean		Flag whether a control plane should be deployed	false
hasTestService	Boolean		Flag whether a test HTTP data plane should be deployed with a test service	false
hasDebugLogging	Boolean		Flag whether to enable debug logging for the services	false
issuer	Boolean		Flag whether this participant is an issuer of credentials. Should be used only for a dataspace authority	false
credentialSubject	Object		Credential subject data for participants
preAuthorizedCode	String		Pre-authorized code for requesting a credential via OID4VCI. Use only when also deploying a dataspace authority or when a code is received from the issuer
config	Map<String, Object>		Custom configuration for the TSG components (controlPlane, ssoBridge, wallet). See also Providing Custom Configuration
overrides	Map<String, Object>		Custom configuration for the Helm charts. See also Providing Custom Configuration
dataPlanes	Map<String, DataPlane>		Data planes that will be configured for this participant

DataPlane

Name	Data Type	Required	Explanation	Default
type	String		Data plane type, only required when type does not match key
tsgDataPlane	Boolean		Flag whether data plane configuration following the TSG data planes should be generated	true
postgres	Boolean		Flag whether a postgres database and credentials should be made	true
subPath	String		Subpath to use for this data plane if different from key and if participant routing is "path"
dnsPrefix	String		DNS prefix to use for this data plane if different from key and if participant routing is "subdomain"
config	Object		Specific data plane config for TSG-based data planes. See also Providing Custom Configuration
overrides	Object		Configuration overrides for the Helm chart values. See also Providing Custom Configuration

PostgreSQL Deployment Options

The CLI uses CloudNativePG for PostgreSQL management with declarative database creation, providing high availability, automated backups, and advanced management features.

Database Management

Databases are created declaratively using CloudNativePG's Database CRD:

• Each application (SSO Bridge, Wallet, Control Plane, Data Planes) gets its own database
• Databases are created as separate Kubernetes resources after the cluster is deployed
• In per-namespace mode, databases are prefixed with participant ID to avoid collisions
• Database lifecycle can be managed independently from the cluster

Deployment Modes

Per-Namespace Mode (default):

• Single shared PostgreSQL cluster for the entire namespace
• All participants share the same cluster with separate databases
• More resource efficient for many participants
• Simplified cluster management with high availability (1 instance by default)
• Configure instances and storage globally in general section

general:
  postgresDeploymentMode: per-namespace
  postgresInstances: 1        # Number of instances for high availability
  postgresStorageSize: "1Gi"  # Storage size for the shared cluster

Per-Participant Mode:

• Each participant gets its own PostgreSQL cluster
• Isolated database resources per application per participant
• Better security isolation
• Easier to manage individual participant resources
• Configure instances and storage per participant using overrides.postgres

general:
  postgresDeploymentMode: per-participant
participants:
  - id: alfa
    name: Alfa
    # ... other participant config
    overrides:
      postgres:
        instances: 2        # Number of PostgreSQL instances for this participant
        storageSize: "5Gi"  # Storage size for this participant's cluster

PostgreSQL Configuration Options

Shared Cluster Configuration (per-namespace mode):

• general.postgresInstances: Number of PostgreSQL instances (default: 1)
• general.postgresStorageSize: Storage size per instance (default: "1Gi")

Per-Participant Cluster Configuration (per-participant mode):

• participant.overrides.postgres.instances: Number of instances for this participant's cluster (default: 1)
• participant.overrides.postgres.storageSize: Storage size for this participant's cluster (default: "1Gi")

Example with mixed configurations:

general:
  namespace: tsg-ecosystem
  postgresDeploymentMode: per-participant
  # ... other general config

participants:
  - id: alfa
    name: Alfa
    # Uses default: 1 instance, 1Gi storage
  
  - id: bravo
    name: Bravo
    overrides:
      postgres:
        instances: 1
        storageSize: "10Gi"  # Larger storage for this participant
  
  - id: charlie
    name: Charlie
    overrides:
      postgres:
        instances: 3        # High availability setup
        storageSize: "5Gi"

CloudNativePG Operator

The CloudNativePG operator must be installed in your cluster. The CLI will automatically check for it and offer to install it if not found:

# The CLI will prompt to install if not found, or install manually:
helm repo add cnpg https://cloudnative-pg.github.io/charts
helm upgrade --install cnpg cnpg/cloudnative-pg --namespace cnpg-system --create-namespace

Providing Custom Configuration

To provide custom configuration for TSG components, you can use the config and overrides fields in the participant configuration.

• The config field allows you to specify settings for the control plane, SSO bridge, and wallet components.
• The overrides field lets you customize any value in Helm chart values.

Both fields can be used together to achieve the desired configuration for your participant.

The CLI will use a deep merge of the properties, allowing to override any default values.

For example, when deploying a participant with a custom control plane configuration, you can specify it as follows:

...
participant:
  ...
  config:
    controlPlane:
      runtime:
        controlPlaneInteractions: manual
  overrides:
    controlPlane:
      ingress:
        class: custom-ingress-class

This will create a control plane that has the runtime setting set via the config property, and the ingress class set via the overrides property.

Note: By default when using arrays in either config or overrides, the CLI will replace the entire array. To append to an existing array, add a "merge-array" as first item in the array. This will append the items to the existing array instead of replacing it.
For example, with the original config:

config:
  some_array:
    - item1
    - item2

And the following override config:

config:
  some_array:
    - "merge-array"
    - item3
    - item4

Results in the final config:

config:
  some_array:
    - item1
    - item2
    - item3
    - item4

Troubleshooting

Common Configuration Errors

Invalid namespace format:

general:
  namespace: "my-namespace"  # ✅ Valid: lowercase with hyphens
  namespace: "My Namespace"  # ❌ Invalid: spaces and uppercase

Missing required fields for issuers:

participant:
  issuer: true
  document: {...}  # ✅ Required when issuer: true
  schema: {...}    # ✅ Required when issuer: true

For more troubleshooting help, see the CLI README and Getting Started Guide.