Building a Raspberry Pi internet speed checker with NodeJS and GoogleDocs

March 18, 2017 0 Comments nodejs, RaspberryPi, speed-test

I frequently experience periods of internet drop outs and wanted to build something to monitor my internet without any manual intervention. I wanted this to create an automated report for me so I could then email it off to my ISP and complain.

So I set about looking on the internet for something that would do this. There are some options and there are a lot of home baked solutions in many different languages.

After a while I remembered I had a Raspberry Pi gathering dust so I dug it out and set off.

Here's how I did it...

Things you'll need:

  • Raspberry Pi
  • SD card
  • Ethernet cable
  • Google account
  • IFTTT account

Setting up your Raspberry Pi

First of all, install Raspbian or some other linux disto on your Raspberry Pi. The simplist way to do this is using the NOOBS method which is described in detail here. --> NOOBS

Make sure you've connected your Raspberry Pi to your router via ethernet cable to get the best results.

Then make sure you've got NodeJS installed on the Pi.

Open the terminal and type

$ node --version

If node is installed you should get a result with a version number similar to below.


If you don't get a version number back, you can download and install NodeJs from

Set up IFTTT trigger

The results of each speed tests need to be pushed into the cloud, the easiest place for the results to end up is a google docs spreadsheet. We can leverage IFTTT's Maker Webhooks channel to get an endpoint which we can post to which will insert the results into a google doc.

You should go to and create a trigger and configure it to add a row to a spreadsheet.

    Maker Event "speed-test", 
    add row to spreadsheet in’s Google Drive

This will create a unique key that looks something like aslkjfadnadsknadsjknad. Keep note of this, we'll use it soon.

Write your node app

So now we've got the Raspberry Pi set up. We have to write the program which will run on a specified interval to do the speed test.

We want this program to do 3 things.

  • Run on a specified interval
  • Do a internet speed test
  • Post results to the cloud

I'm going to use the node-cron package to run the speed test on a specified interval. For the speed test, I'm going to use speedtest-net.

Now, create a node application, and install the dependencies you'll need.
(I've included isomorphic-fetch so we can easily make a request to our IFTTT endpoint)

$ mkdir node-speed-checker
$ cd node-speed-checker
$ npm init -y
$ npm install node-cron speedtest-net isomorphic-fetch

Lets create a function that runs a speed test and then plucks the correct information from the returned data. We then want to create a querystring that we can append to our IFTTT trigger url. I've added an environment variable here called IFTTT_KEY so that we can push the key in when we run up the app.

function runSpeedTest() {  
    speedTest({maxTime: 5000})
    .on('data', data => {
        const meta = {
            client: data.client,
            server: data.server
        const queryString = `value1=${}&value2=${data.speeds.upload}&value3=${JSON.stringify(meta)}`;
        const key = process.env.IFTTT_KEY;
    }).on('error', err => {

Now that we have our function that will run the speed test, we need to set up the scheduler.

cron.schedule('0 * * * *', runSpeedTest);  

This will run runSpeedTest every hour on the hour.

You can check out this full code at

Now, you just need to put your code onto your Raspberry Pi and run the app. The easiest way to do this is to push your code to a github repo, SSH into the Pi and clone the codebase.

When you've done all this this, install the dependencies, set the IFTTT_KEY environment variable and run the app.

$ git clone
$ cd node-cron-speed-checker
$ npm install
$ IFTTT_KEY=aslkjfadnadsknadsjknad node index.js

This will now run a speed-test from your Raspberry Pi and log the results to your Google Spreadsheet. After a few days, you'll start to build up a result set and over a few weeks you'll have enough data to create some graphs and see your peak and low times.