Chapter 3: Portal and Observability
Your customers are not API-first developers. They need a web interface. And you need to know when something breaks.
┌──────────────────────────────────────────────────┐
│ Your Platform (CSP) │
│ │
│ ┌────────────────────────────────────────────┐ │
│ │ User Portal (CSP) │ │
│ │ Calls Management API + Identity API │ │
│ └────────────────────────────────────────────┘ │
│ │
│ ┌────────────────────────────────────────────┐ │
│ │ Monitoring & Observability (CSP) │ │
│ │ Metrics · Logs · Alerts · Health checks │ │
│ └────────────────────────────────────────────┘ │
│ │
│ ┌────────────────────────────────────────────┐ │
│ │ CFM + EDC-V + Data Plane │ │
│ │ (same as baseline) │ │
│ └────────────────────────────────────────────┘ │
│ │
│ ┌────────────────────────────────────────────┐ │
│ │ Infrastructure (CSP) │ │
│ │ (same as baseline) │ │
│ └────────────────────────────────────────────┘ │
└──────────────────────────────────────────────────┘
What You Build at This Level
| Component | What it takes |
|---|---|
| User Portal | A web application for your customers. Catalog browsing, contract management, credential viewing, transfer status. The portal calls the Management API and Identity API on behalf of the logged-in user. Authentication via your IDP. |
| Monitoring & Observability | Health checks against component endpoints, metrics collection (Prometheus or equivalent), log aggregation, alerting. Standard cloud-native observability — nothing dataspace-specific. |
The Observability API on each EDC component exposes health and readiness endpoints — you just need to scrape and alert on them.
Next: Customer Applications