Last month Alex Markessinis released a very nice project based on balenaSanse to manage the temperature and humidity of a data center and upload the data to an external InfluxDB database.
Follow along as we go through the process of setting up this project to capture data from remote sensors and send the data to a central InfluxDB database with the help of a Raspberry Pi and balenaCloud.
Table of contents
- Hardware required
- Software required
- Setting up the Raspberry Pi
- Enable the 1-Wire interface
- Raspberry Pi GPIO Pinout
- Create a Device Image and Flash it
- Deploy the Project
- Configure your Application
- Viewing the Dashboard
- Conclusion
Hardware required
- microSD card -16GB recommended
- Raspberry Pi 3 B+
- DHT22/AM2302 Digital Temperature And Humidity Sensor Module (Amazon, AliExpress)
Software required
- A balenaCloud account
- An InfluxDB server – setting up this is outside the scope of this guide, but check their installation guide
- (Optional) Grafana dashboard to display information from InfluxDB
Setting up the Raspberry Pi
After setting up an account on balenaCloud, create a new application called PiProbes. Select Raspberry Pi 3 for the device type and leave the project type as Starter.
Enable the 1-Wire interface
Once your application is created, you need to enable the 1-Wire Interface on your device. To do so, go to Fleet Configuration in your balenaCloud account and add a new custom variable called RESIN_HOST_CONFIG_dtoverlay with the value of w1-gpio.
Raspberry Pi GPIO Pinout
Below is a diagram of the GPIO Pinout for the Raspberry Pi:
Install the probe on the Pi using pins:
| Pin | Function |
|---|---|
| Pin 1 | (+) Power 3V3 |
| Pin 7 (GPIO 4) | GPIO 1-Wire Data Out |
| Pin 9 (GND) | (-) GND |
This is the pinout configuration used for this project. Feel free to change it up as required, just be sure to let the piProbe application know which pin you’re using for the 1-Wire interface by changing the configuration variables (shown later in this guide) appropriately.
Below are some images of the probe wired up:
Create a Device Image and Flash it
The next step is to create a device image. Go to Devices in your balenaCloud and click on the + Add Device button, select either Development or Production image type. If you’re planning to develop the project further and need local development access via balena push and SSH use the Development image type.
If you wish to connect to the internet via WiFi, enter the credentials and then click on the Download balenaOS button to save the image file to your computer.
After the image has finished downloading, open balenaEtcher, select the image zip file, ensure the right SD card is inserted into your computer and selected, and finally, flash it.
Once flashing is completed, insert the SD card into your Pi and turn it on. You’ll see it appear in the devices section of your dashboard after a couple of minutes.
Deploy the Project
To deploy the project to your device, you first need to have the balena-cli installed on your computer, you can check the balena CLI Installation Instructions here.
The first time you use the balena-cli, you will be asked to authenticate to balenaCloud, which you can do by typing balena login (You only need to do this the first time you use the CLI).
Now go to https://github.com/MelonSmasher/piProbe and download or clone the repository to your machine. From within the newly created directory, type balena push PiProbes to push the code to your device.
git clone https://github.com/MelonSmasher/piProbe.git && cd piProbe; balena push PiProbes 
Configure your Application
Go to the balena dashboard and configure your Fleet Service Variables with the following variables (Note that you will be pointing it to your external InfluxDB database hosted on another server):
Fleet Service Variables
| Service | Name | Required | Default Value | Valid Values | Description |
|---|---|---|---|---|---|
| piProbe | INFLUXDB_HOST | Yes | “” | STRING | The ip or hostname of your InfluxDB server |
| piProbe | INFLUXDB_PORT | No | 8086 | INT | The port InfluxDB is listening on |
| piProbe | INFLUXDB_USER | No | “” | STRING | The InfluxDB user used to authenticate |
| piProbe | INFLUXDB_PASSWORD | No | “” | STRING | The InfluxDB user password |
| piProbe | INFLUXDB_DB | Yes | False | STRING | The db name where the measurements will be stored |
| piProbe | INFLUXDB_INTERVAL | No | 10 | INT | The interval to measure data and write to InfluxDB |
| piProbe | INFLUXDB_SSL | No | False | STRING: Yes/No | Should the client use ssl to connect to InfluxDB? |
| piProbe | INFLUXDB_SSL_VERIFY | No | False | STRING: Yes/No | Should the client verify ssl to certs on InfluxDB? |
| piProbe | GPIO_PIN | No | 4 | INT | The GPIO 1-Wire pin |
| piProbe | GPIO_SENSOR | Yes | “” | STRING: DHT11/DHT22/AM2302 | The sensor model |
Next, navigate back to your devices, select your Pi, and click on ‘Service Variables’. From here you’ll fill out the INFLUXDB_LOCATION_TAG for this device. This lets you know where the device is physically located when you query the database (The value can be any string).
Device Service Variables
| Service | Name | Required | Default Value | Valid Values | Description |
|---|---|---|---|---|---|
| piProbe | INFLUXDB_LOCATION_TAG | Yes | “” | STRING | Where is this probe? `living room`, `the moon`, `Winterfell`? Fill this out so you know where the probe is. |
After adding the variables, the piProbe service will restart automatically on your device and will begin reporting to your InfluxDB server!
Now with the sensor data being sent to your InfluxDB database, you can then setup a Grafana dashboard to display all the information nicely! Some examples of the dashboards that Alex set up are below:
Conclusion
In this post, we saw how you can easily build a temperature and humidity sensor using a Raspberry Pi and balenaCloud, capture the readings, and send them to a central InfluxDB database for reporting and visualisation.
If this post has raised any questions or you’d like to know more, please join us in our forums, on Twitter @balena_io, on Instagram @balena_io or on Facebook where we’d be more than happy to answer!
