Skip to content

Latest commit

 

History

History
161 lines (116 loc) · 5.88 KB

balena-fleet-deployment.md

File metadata and controls

161 lines (116 loc) · 5.88 KB
Raw

Deploying on a Balena fleet

You can use the Balena disk images provided in the releases page to install Anthias in your device. However, if you want more control and transparency over the Balena deployment, follow through the steps below to deploy Anthias on your Balena fleet.

Getting started

  • Create a balenaCloud account if you don't have one already.
  • Install the balena CLI on your computer. It's recommended to install it on a Linux machine but it should work on Windows and macOS as well. If you're on Windows, you can use the Windows Subsystem for Linux (WSL) to install the CLI. Click here for the installation instructions.
  • On your terminal, login to your balenaCloud account using the CLI by running balena login. You'll be prompted to select a login method. Select Web authorization, which is the recommended way to login. The browser will open a new tab. Click Authorize to proceed.
  • Install balenaEtcher on your computer. You can download it from here.

Create and configure a new fleet

Open your browser and go to https://dashboard.balena-cloud.com. Login to your balenaCloud account if you haven't already. You should be redirected to the dashboard.

balena-ss-01

Click on the Create fleet button. Give your fleet a name and select the appropriate device type. Click on the Create new fleet button. You should be redirected to the fleet's summary page.

balena-ss-02

balena-ss-03

We'll be doing the initial fleet configuration via CLI. Open your terminal and run the following commands:

$ balena env add RESIN_HOST_CONFIG_gpu_mem $GPU_MEM_VALUE --fleet $FLEET_NAME

# Run the command below only if you're using a Raspberry Pi 4, as it uses
# VLC for video playback.
$ balena env add RESIN_HOST_CONFIG_dtoverlay vc4-fkms-v3d --fleet $FLEET_NAME

Replace $GPU_MEM_VALUE with the GPU memory value you want to use, as long as it's less than the total memory of your device. For example, for a 4GB Raspberry Pi 4, you can use 256 as the GPU memory value. Having insufficient GPU memory might cause video playback issues.

You can confirm that the changes went through by running the following command:

balena envs --fleet $FLEET_NAME --config

Here's a sample output:

ID      NAME                           VALUE        FLEET
1979572 RESIN_HOST_CONFIG_dtoverlay    vc4-fkms-v3d gh_nicomiguelino/anthias-pi4
1979571 RESIN_HOST_CONFIG_gpu_mem      1024         gh_nicomiguelino/anthias-pi4

Alternatively, you can check the releases page of that fleet and look for the RESIN_HOST_CONFIG_gpu_mem and RESIN_HOST_CONFIG_dtoverlay variables.

balena-ss-04

Deploy changes to the fleet

Before proceeding, make sure that you have logged in by running balena login. You can verify if you're logged in by running balena whoami.

Open your terminal and clone the Anthias repository if you haven't already:

$ cd $WORKSPACE_DIRECTORY
$ git clone [email protected]:Screenly/Anthias.git
$ cd Anthias/

Run the following command:

$ ./bin/deploy_to_balena.sh \
    --board $BOARD_TYPE \
    --fleet $FLEET_NAME \

$BOARD_TYPE could be any one of the following — pi1, pi2, pi3, pi4. You can run ./bin/deploy_to_balena.sh --help for details.

Running the command above will pull the latest Docker images from Docker Hub and push them to your balenaCloud account. It will also create a new release and deploy it to your fleet.

If you want to deploy your local changes, run the following command instead:

# Take note of the --dev flag.
$ ./bin/deploy_to_balena.sh \
    --board $BOARD_TYPE \
    --fleet $FLEET_NAME \
    --dev

You can also includa a --shm-size flag to specify the shared memory size, which defaults to 256mb. For example:

$ ./bin/deploy_to_balena.sh \
    --board $BOARD_TYPE \
    --fleet $FLEET_NAME \
    --shm-size 512mb

It would take a while for the deployment to finish. Once it's done, you should see the new release in the fleet's summary page. You can now add your devices to the fleet and they should be able to download the new release.

balena-ss-05

balena-ss-06

Add a new device to the fleet

Insert a microSD card to your computer.

Open your browser and go to https://dashboard.balena-cloud.com. You should be redirected to your balenaCloud dashboard. Click on the fleet you created earlier.

Click "Add device" on the fleet's summary page. The "Add new device" page should appear. Leave the default values unchanged. Click "Flash" when ready.

balena-ss-07

A new browser tab will open prompting you to open balenaEtcher. Click "Open" to proceed. Select the microSD card you inserted earlier and click "Flash". Wait for the flashing process to finish.

balena-ss-08

Remove the microSD card from your computer and insert it to your device. Power on the device and wait for it to appear on the fleet's summary page. Once it appears, click on the device's name to go to its summary page. Be patient while balenaCloud downloads the Docker images and starts the containers. It might take a while for the device to appear online.

balena-ss-09

balena-ss-10

Once done, the display should show the Anthias splash screen. You can now add assets via the web interface.