Getting started with Observability

Create in the console

1. Open the Thalassa Cloud Console.
2. Navigate to Observability → Workspaces.
3. Click Create Workspace.
4. Configure:
   • Name — for example production-observability
   • Region — usually the same region as your workloads
   • Retention — days to retain metrics and logs (optional; regional default if omitted)
5. Click Create and wait until status is Ready.

Create via API

curl -X POST "https://api.thalassa.cloud/v1/observability/workspaces" \
  -H "Authorization: Bearer $TOKEN" \
  -H "Content-Type: application/json" \
  -H "X-Project-Identity: $PROJECT_ID" \
  -d '{
    "name": "production-observability",
    "regionIdentity": "nl-01",
    "retentionDays": 30
  }'

Creating a workspace enables both metrics and logs backends. Endpoints appear on the resource when status is ready.

Workspace endpoints

On the workspace Overview, copy:

URLs look like:

https://prometheus.<region>.thalassa.cloud/workspace/obsw-…/api/v1/push
https://loki.<region>.thalassa.cloud/workspace/obsw-…/loki/api/v1/push

The workspace ID is already in the path — use the URL as provided.

Authentication

Use a service account for collectors:

1. Create a service account under IAM → Service Accounts.
2. Bind policies that allow workspace data-plane access — at minimum Prometheus remote write for metrics ingest (for example prometheus:RemoteWriteFullAccess or prometheus:FullAccess), scoped so the principal can use this workspace.
3. Create OIDC client credentials for the service account.
4. Ensure the token includes the required scopes (for example observability:prometheus:remote_write).

Personal access tokens work for testing but are not recommended for production agents.

prometheus.yml

Replace the URL with your workspace remote write URL from the console:

global:
  scrape_interval: 15s
  external_labels:
    cluster: production
    region: nl-01

scrape_configs:
  - job_name: "kubernetes-pods"
    kubernetes_sd_configs:
      - role: pod
    relabel_configs:
      - source_labels: [__meta_kubernetes_pod_annotation_prometheus_io_scrape]
        action: keep
        regex: true

remote_write:
  - url: https://prometheus.<region>.thalassa.cloud/workspace/obsw-<id>/api/v1/push
    oauth2:
      client_id: <service-account-client-id>
      client_secret: <service-account-client-secret>
      token_url: https://api.thalassa.cloud/oidc/token
    queue_config:
      max_samples_per_send: 1000
      batch_send_deadline: 5s
      max_retries: 3

Reload Prometheus

# systemd
sudo systemctl reload prometheus

# or API
curl -X POST http://localhost:9090/-/reload

For Prometheus Operator and advanced queue/relabel settings, see Metrics remote write.

Check the agent

1. Open your Prometheus UI → Status → Runtime (remote write section).
2. Confirm the remote write endpoint is up and samples are sent without sustained errors.

Query from the gateway

curl -H "Authorization: Bearer $TOKEN" \
  "https://prometheus.<region>.thalassa.cloud/workspace/obsw-<id>/api/v1/query?query=up"

Query in the console

1. Open the workspace.
2. Open the Metrics view.
3. Run up and confirm series from your collector.