Getting Started with ClusterNest Managed Prometheus

ClusterNest Managed Prometheus runs Cortex behind a Prometheus-compatible remote-write/query API, so any tool that speaks the Prometheus HTTP API (Prometheus itself, Grafana Agent, an OpenTelemetry Collector, Grafana) works against it without changes. This guide creates a cluster, fetches its credentials, and runs a first query.

Every example below needs your organization's numeric ID - find it in the console (it's in the URL once you're inside an organization), or via GET /auth/user-info with a bearer token, which lists every organization you belong to along with its id.

1. Create a cluster

Console

• Login/Signup for an account.
• Click "Launch Cluster", choose Prometheus, pick a tier and a retention period.
• Click "Launch". The cluster is ready once its state shows available.

Terraform

resource "clusternest_prometheus" "metrics" {
  name            = "metrics"
  tier            = "standard"
  organization_id = 123
}

See the resource reference for the full field list.

Raw API

POST /cluster/prometheus/, authenticated with a bearer token (see Get an access token):

curl -X POST https://api.clusternest.com/cluster/prometheus/ \
  -H "Authorization: Bearer $ACCESS_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "metrics",
    "organization_id": 123,
    "tier": "standard"
  }'

The trailing slash on /cluster/prometheus/ is required - without it the API 307-redirects, and most HTTP clients don't resend a POST body across that automatically.

The response includes the cluster's id - every other call in this guide needs it.

2. Get your credentials

Every Prometheus cluster has three static credentials: admin (full access), readonly (query only), and write (push only) - see Managing credentials for exactly what each one can reach.

Console

Shown automatically once the cluster in step 1 finishes creating - save them then. The console has no way to view them again later; use Terraform or the raw API for that.

Terraform

data "clusternest_prometheus_credentials" "metrics" {
  cluster_id = clusternest_prometheus.metrics.id
}

Raw API

curl https://api.clusternest.com/cluster/prometheus/$CLUSTER_ID/credentials \
  -H "Authorization: Bearer $ACCESS_TOKEN"

{
  "admin": "<password>",
  "readonly": "<password>",
  "write": "<password>"
}

3. Run your first query

Fetch the cluster's hostname (the prometheus_url field, returned by GET /cluster/prometheus/$CLUSTER_ID or shown in the console, minus the https:// - that's $PROMETHEUS_HOST in every other guide here too), then query it with PromQL under the /prometheus prefix, using the admin username and the admin password from the step above:

curl -u "admin:$ADMIN_PASSWORD" \
  "https://$PROMETHEUS_HOST/prometheus/api/v1/query?query=up"

An empty result array is expected here - nothing has pushed metrics into the cluster yet.

Next steps

• Push metrics in with the write credential.
• Query and wire up Grafana with the readonly credential.
• Let your own identity provider authenticate directly instead of managing static passwords.