Observability
Observability
The complete MQTT platform in one small binary: broker, SQL rules, data connectors and multi-tenancy in a single process.
Read broker counters. Use it to watch traffic and drops. You get Prometheus text from the metrics endpoint.
flowchart LR Op["operator"] -- "GET /api/v1/metrics" --> API["management API"] API -- "read indramqtt_ counters" --> R["router"]Before you start
- Start the broker from First start.
- Set
INDRA_API_KEYto an operator key. The examples read it.
Note: Calls except
/healthzneed an operator API key. Pass it asAuthorization: Bearer $INDRA_API_KEY.
Live throughput, latency, and Prometheus metrics
GET /api/v1/metrics
Get broker metrics
Operation id: getMetrics.
Parameters:
No parameters.
Request body:
No request body.
Responses:
| Status | Description |
|---|---|
| 200 | Prometheus text exposition of live counters |
Response 200 body:
Media type text/plain:
{ "type": "string"}Notes:
- Related calls sit on the same tag page.
Verified: the example flow below runs this call.
Example:
# Illustrative request. Values come from the schema above.# The example flow below runs verified calls on a live broker.curl -s -X GET 'http://127.0.0.1:18083/api/v1/metrics' \ -H "Authorization: Bearer $INDRA_API_KEY"Example flow
Read real rows through the management API, then read them back. The docs harness runs this flow against a started broker.
python3 - <<'PY'import os, urllib.requestport = os.environ.get("API_PORT", "18083")key = os.environ.get("API_KEY", "")base = f"http://127.0.0.1:{port}"req = urllib.request.Request(base + "/api/v1/metrics", method="GET")req.add_header("Authorization", f"Bearer {key}")with urllib.request.urlopen(req, timeout=5) as r: body = r.read().decode() assert r.status == 200, bodyassert "indramqtt_" in body, bodyreq2 = urllib.request.Request(base + "/api/v1/metrics", method="GET")req2.add_header("Authorization", f"Bearer {key}")with urllib.request.urlopen(req2, timeout=5) as r2: body2 = r2.read().decode() assert r2.status == 200, body2assert body == body2, "read back the same metrics text"print("OK")PYOKCheck that it works
Start the broker with this file. The health endpoint returns OK.
[node]id = "doc-ref"
[listeners.tcp]bind = "0.0.0.0:1883"
[listeners.tls]bind = "0.0.0.0:8883"
[listeners.ws]bind = "0.0.0.0:8083"
[listeners.wss]bind = "0.0.0.0:8084"
[listeners.api]bind = "0.0.0.0:18083"
[logging]level = "warn"python3 - <<'PY'import os, socketport = int(os.environ.get("API_PORT", "18083"))sock = socket.create_connection(("127.0.0.1", port), timeout=5)sock.sendall(b"GET /healthz HTTP/1.0\r\nHost: docs\r\n\r\n")print(sock.makefile().read().strip().splitlines()[-1])PYOKCheck one documented call. GET /api/v1/metrics answers 200 with its rows. It prints OK.
python3 - <<'PY'import os, urllib.requestport = os.environ.get("API_PORT", "18083")key = os.environ.get("API_KEY", "")base = f"http://127.0.0.1:{port}"req = urllib.request.Request(base + "/api/v1/metrics", method="GET")req.add_header("Authorization", f"Bearer {key}")with urllib.request.urlopen(req, timeout=5) as r: body = r.read().decode() assert r.status == 200, bodyassert "indramqtt_" in body, bodyprint("OK")PYOKNext steps
- API reference: the other tags.
- Configuration reference: the settings behind the rows.
- Metrics reference: the counter names.