Quick start

This exercise takes you from zero: create a workspace, register one device, send data over MQTT, store it with a rule, and show it on a dashboard. You need a Celestian account, the MQTT broker address from your administrator, and mosquitto_pub (or Python/Node.js, see the code tabs).

Tip

No device yet? Use the celestian-sim simulator to send realistic fake sensor data.

1. Sign in or sign up

   Open the Celestian address and choose Sign up (if registration is open) or Sign in. Verify your e-mail address if asked. If registration is closed, ask your administrator for an invitation (see Invitations).

   

2. Create a workspace

   On the Workspaces page, choose Create workspace. Fill in Name (for example Cold Storage Nusantara) and Route (for example cold-storage). The route is used in MQTT topics and cannot be changed later.

   

3. Create a channel

   Menu Channels → Create channel. Enter the name Warehouse Telemetry and the route warehouse-telemetry.

   

4. Create a device and save its key

   Menu Devices → Create device. Enter the name Freezer Room A1 Sensor and the Serial SN-A1. After saving, Celestian shows the device key only once. Copy and store it. The device ID (a UUID) is visible on the detail page.

   Caution

   The device key is stored as a hash and cannot be shown again. If you lose it, create a new key on the device detail page.

5. Connect the device to the channel

   Open the channel detail, the Connections section, choose Connect, pick the device, and tick the Publish type. Without a publish connection the broker drops a device that tries to send.

6. Add a storage rule

   In the channel detail (the Settings section), the Storage card shows the Not stored flag. Choose the Store in Internal DB (SenML) button. The rule it creates is Channel Subscriber → Bypass → Internal DB (SenML).

7. Send data from the device

   mosquitto_pub

   DEV=<device-id>
   mosquitto_pub -h <broker-address> -p 1883 -V mqttv311 -q 1 \
     -i "$DEV" -u "$DEV" -P '<device-key>' \
     -t 'm/cold-storage/c/warehouse-telemetry/room-a1' \
     -m '[{"n":"temp_inlet","u":"Cel","v":22.4},{"n":"humidity","u":"%RH","v":48.3},{"n":"door_open","vb":false}]'

   Python

   import json
   import paho.mqtt.client as mqtt
   
   DEV, KEY = "<device-id>", "<device-key>"
   c = mqtt.Client(mqtt.CallbackAPIVersion.VERSION2, client_id=DEV)
   c.username_pw_set(DEV, KEY)
   c.connect("<broker-address>", 1883)
   c.loop_start()
   payload = json.dumps([{"n": "temp_inlet", "u": "Cel", "v": 22.4}])
   c.publish("m/cold-storage/c/warehouse-telemetry/room-a1", payload, qos=1).wait_for_publish()
   c.loop_stop()

   Node.js

   import mqtt from "mqtt"; // npm i mqtt
   const c = mqtt.connect("mqtt://<broker-address>:1883", {
     clientId: "<device-id>", username: "<device-id>", password: "<device-key>",
   });
   c.on("connect", () =>
     c.publish("m/cold-storage/c/warehouse-telemetry/room-a1",
       JSON.stringify([{ n: "temp_inlet", u: "Cel", v: 22.4 }])));

   The Credentials tab on the device page generates the same examples with the ID and topic already filled in.

8. See the messages in Messages

   Open Messages, pick the channel, and turn on Live to watch messages arrive instantly. Stored messages appear in the table.

   

   Stored messages in Messages.

9. Create a dashboard

   Menu Dashboards → Create dashboard, then Edit mode → Add widget (for example Line Chart). Pick the channel, the value name temp_inlet, and a time window.

   

   The end result: a monitoring dashboard.

Next steps

• Understand the payload formats so your data is more compact and structured.
• Add an alarm for when the temperature crosses a limit.
• Set up members and roles so your colleagues can join.
• If you use a gateway, read Gateways and child sensors.