#RaspberryPi – Deploy Azure Blob Storage on IoT Edge, lessons learned #AzureIoT

Hi !

I’ve been using Azure Blob Storage for a very long time. I won’t share details so you won’t laugh a lot. And now, in the amazing Azure IoT world, we have the chance to use a custom module in out Edge devices. As the documentation mention, this module is useful in scenarios:

where data needs to be stored locally until it can be processed or transferred to the cloud. This data can be videos, images, finance data, hospital data, or any other unstructured data.
when devices are located in a place with limited connectivity.
when you want to efficiently process the data locally to get low latency access to the data, such that you can respond to emergencies as quickly as possible.
when you want to reduce bandwidth costs and avoid transferring terabytes of data to the cloud. You can process the data locally and send only the processed data to the cloud.

In the references section I shared the official documentation. So, while you are reading the doc and following the steps, I’ll share some insights.

Install module from the Marketplace

As usual, we can deploy the module using a custom JSON definition. For 1-to-1 devices, install the module from the market place is a quick option. Just go your IoT Edge device panel, select [Set Modules], and in the IoT Edge Modules section, select [+ Marketplace Module].

azure iot install module from marketplace

On the search panel, search for “storage” and check the results.

azure iot install module from marketplace azure blob storage on Iot edge

2 clicks later, the module is in your definition. However, some config is needed. I’ll go on this for next section.

Settings / Config for the Azure Blob Storage on IoT Edge module

Time for module configuration. In the official documentation describes the 2 main sections to work on.

The main values that we will need are these ones, so you may need to create / define them in advance.

Storage account name. Account names should be 3 to 24 characters long, with lowercase letters and numbers. No spaces.
Storage account key. Replace with a 64-byte base64 key. You can generate a key with tools like GeneratePlus. You’ll use these credentials to access the blob storage from other modules.

Environment Variables

You need to complete these values in the [Environment Variables] section.

Module Name needs to be lowecase

At some point the documentation mention / suggest to change the name to lowercase. Please do this, I had several issues on my Raspberry Pi, and somehow a lowercase name fixed some deployment issues.

Container Create Options

This section is mostly straight forward. I used the Account Name and Account Key defined in previous steps and I also created a specific folder to store the blob [/home/pi/azurestorageiot]. Again, here the documentation makes everything easy to configure.

{
  "Env":[
    "LOCAL_STORAGE_ACCOUNT_NAME=<your storage account name>",
    "LOCAL_STORAGE_ACCOUNT_KEY=<your storage account key>"
  ],
  "HostConfig":{
    "Binds":[
            "/home/pi/azurestorageiot"
    ],
    "PortBindings":{
      "11002/tcp":[{"HostPort":"11002"}]
    }
  }
}

Module Twin Settings options

Main remark here: while you are in testing mode, set the uploadOn to false (line 8). Otherwise the module will

be installed with default options and no containers
Search for a local container [cont1] and will log tons of errors, because the container does not exist

And the lines 11 to 15 were also tricky to understand. I was a expecting a source / target pair to define source and target containers, however the definition works different.

For each source container (the ones in our IoT Edge device), we need to create a section (lines 12 to 14) with the name of the IoT Edge container (cont1), and the target definition with the name of the Azure container (line 13).

Again, I spend sometime here, mostly because I was with a different mindset before understanding how this works.

And that’s it! I deployed and tested this module in a Raspberry Pi and it works like a charm!

References

Happy coding!

Greetings

El Bruno

My posts on Raspberry Pi ⚡🐲⚡

Dev posts for Raspberry Pi

Tools and Apps for Raspberry Pi

Setup the device

Hardware

¿Con ganas de ponerte al día?

En Lemoncode te ofrecemos formación online impartida por profesionales que se baten el cobre en consultoría:

Si tienes ganas de ponerte al día con Front End (ES6, Typescript, React, Angular, Vuejs…) te recomendamos nuestros Máster Front End: https://lemoncode.net/master-frontend#inicio-banner
Si te quieres poner al día en Backend (stacks .net y nodejs), te aconsejamos nuestro Bootcamp Backend: https://lemoncode.net/bootcamp-backend#bootcamp-backend/banner
Y si tienes ganas de meterte con Docker, Kubernetes, CI/CD…, tenemos nuestro Bootcamp Devops: https://lemoncode.net/bootcamp-devops#bootcamp-devops/inicio