Table of contents:
- Deploy Smart Mirror AWS cloud components
- Deploy AWS Cloudformation Stacks
- Validate skill in Amazon developer console and create product
- Check skill on your Alexa App (in your phone)
- Add Smart Mirror New Product in amazon developer console
- Configure Systems Manager Agent for your Raspberry Pi
- Add Deployment step for Raspberry Pi in the CodePipeline - in AWS Console
- Open AWS CodeStar smart-mirror project and edit the smart-mirror-pipeline (Through AWS Codepipeline or CodeStar service):
- Configure Alexa Gadget in Raspberry Pi and BT pairing with Echo
Deploy Smart Mirror components and functionality - high level steps:
- Set up CI/CD pipeline and example Alexa skill (AWS CodeStar)
- Connect AWS and Alexa Developer Accounts
- Work with AWS Cloud9 IDE - and deploy Smart Mirror Example code to your repository
- Add necessary AWS Roles and policies (CloudFormation Templates)
- Alexa Skill setup (In Amazon Developer Console)
- Add Alexa Skill - and validate connectivity
- Add Smart Mirror as Product toi Amazon Developer Console
- Deploy Systems Manager Agent to Raspberry Pi - triggers automated CodeDeploy agent installation through Association
- Create Application and Raspberry Pi Smart Mirror SW automated Deployment in AWS CodeDeploy
- Configure Alexa Gadget and BT connectivity
- Installation
DONE - Verify, validate and start to add functionality
NOTE: See additional details Tutorial: Create an Alexa Skill Project in AWS CodeStar
Log in to your AWS Account with Admin privileges and open AWS Console.
NOTE: Validate that region is where you will deploy your smart mirror. In the examples we use Ireland (eu-west-1).
Open AWS CodeStar Service (Through Search Bar - or browsing to service)
Select “Create Project” in AWS CodeStar
Select “Create service role”
Select Templates “Hello World Skill” Python Alexa Skill as project template → Select "Next"
Fill in "Project name" to be "smart-mirror" this will also populate, "Project Id" and "Repository Name" for same value - select CodeCommit as repository - see screenshot below for details
After this select “Connect Amazon developer account” you created for the project - fill in the credentials - "Confirm" OAuth request and "Allow" sign-in
"Allow" and connect account - or log in if not yet done - Select your "Vendor Account"
Finally select "Next" - Review changes and execute selecting "Create project"
NOTE: Project takes few minutes to provision
After a while you have AWS CodeStar project available:
- Repository
- Pipeline
- Other Project Resources
NOTE: We will use AWS Cloud9 IDE - other IDE is also possible and instructed under IDE - or you can use your existing developer machine. See more details about Cloud9 in Cloud9 Tutorial
Select "Set up AWS Cloud9" - in AWS CodeStart Overview -tab
Fill in details for Cloud9 - see example in below picture:
NOTE: You can use default VPC and subnets (Or if you have specific requirements change accordingly)
Select "Create Environment"
Environment creation takes some minutes - and is visible only after environment is created - refresh page to get visibility.
After IDE is ready you can open it from AWS CodeStar “Open IDE” button
You now directly have access to sample repository deployed by CodeStar (python example) in IDE. To make it easier to work with IDE clean up Welcome pages and make terminal bigger - see screenshot above
There is now IDE in place to work with smart-mirror code
Copy smart-mirror.zip from smart-mirror-mvp folder to IDE / repository (you can clone the repo there or use other method of your choice)
Browse to /environment folder on your Cloud 9 IDE Terminal and unzip the "smart-mirror.zip" containing source code for smart mirror - overwriting sample code currently in repository.
To upload files to Cloud9, either drag a local file directly from your desktop into the Cloud9 file tree (in environment folder), or use the File > Upload Local Files menu item
Before pushing you can also configure your git username and password (git config) as per instructions in Cloud9 console.
The default name for the skill that will be created when you push the code is currently "smart mirror a.k.a. unicorn". If you want to change that name you can change it using Cloud9 by editing the name property in skill.json.
Screenshot from extraction:
Commands to unzip the file in Cloud9 IDE terminal:
cd ~/environment/
unzip -o smart-mirror.zip
cd smart-mirror
git add .
git commit -m “First Push”
git push
If you view AWS CodePipeline service (or Pipeline in CodeStar) in AWS console you can see AWS CodePipeline building sample pipeline with smart-mirror code.
CI/CD in Action!
NOTE: Use region where you deployed. In Examples we use eu-west-1 (Ireland)
Additional documentation on CloudFormation CLI: https://docs.aws.amazon.com/AWSCloudFormation/latest/UserGuide/cfn-using-cli.html
Create AWS SSM association with AWS CloudFormation using AWS CLI in Cloud9 IDE with configuration file CodeDeployAssociation.yaml as template body.
Access AWS Cloud9 IDE you created before and there open terminal and browse to ~/environment/smart-mirror/cloudformation-templates folder and execute cloudformation CLI.
Create Stack with commands:
cd ~/environment/smart-mirror/cloudformation-templates/
aws cloudformation create-stack --stack-name SmartMirrorAssociation --template-body file://./CodeDeployAssociation.yaml
Screenshot:
Create AWS SSM association with AWS CloudFormation using AWS CLI in Cloud9 IDE with configuration file SmartMirrorDeviceRole.yaml as template body.
Access AWS Cloud9 IDE you created before and there open terminal and browse to ~/environment/smart-mirror/cloudformation-templates folder and execute cloudformation CLI.
Create Stack with commands:
cd ~/environment/smart-mirror/cloudformation-templates/
aws cloudformation create-stack --stack-name SmartMirrorDeviceRole --template-body file://./SmartMirrorDeviceRole.yaml --capabilities CAPABILITY_NAMED_IAM
Screenshot:
Open AWS Systems Manager Service and select “Hybrid Activations” under Node Management in left pane Select “Create an Activation”
Change activation description - and name as in below screenshot
NOTE: Activation default expiration is 24 hours - so make sure use it during the window or adjust "Activation expiry date" accordingly
Be sure to use SmartMirrorDeviceRole you Created earlier in this procedure (by deploying the CloudFormation stack)
Fill in as above and Select "Create activation"
NOTE: Copy down the details (Activation ID and Code) in safe place!
Activation Code = <your value - record to safe place>
Activation ID = <your value - record to safe place>
Open Amazon Developer Services
Open Developer Console from top right
Open Alexa / Alexa Skills Kit
In Alexa Console you have a skill (created by Smart Mirror code). Skill name is by default "smart mirror a.k.a. unicorn" unless you have changed it before pushing the code.
Enable Development mode (under Test menu)
Test the skill with your invocation word (Find this in sample code skill.json - below we use racecar - default is unicorn).
NOTE: No need to worry about "Smart mirror not found" - HW is not configured or connected yet :)
Open Alexa app in mobile: More → Skills & Games → Choose "Dev"
You should see your skill there provided you are on right account (in below skill is named "mirror test")
NOTE: Make sure the account you are using is the same account you use in your device
Open Amazon Developer Console
See additional details in Alexa Gadgets Raspberry repository (You don't need to follow those instructions. Instructions below are sufficient)
Select "Developer Console" (top right)
Open Alexa / Alexa Voice Service
Select "Manage Your Products"
Select "Add New Product"
Fill in as in example below
And in the end of page press "FINISH"
Select “Edit“ your product (three dots next to name) just created and write down to safe place:
Amazon ID = <your value>
Alexa Gadget Secret = <your value>
NOTE: You will need above those when creating parameters for your device
Open AWS CloudFormation console and use configuration file ParameterStore.yaml from IDE / "cloudformation-template" folder in AWS Console as Admin user
Select Create Stack : With new resources(standard)
Choose: "Template is ready" and "Upload a template file"
Browse to location that has "ParameterStore.yaml" (in IDE "cloudformation-templates" folder) and press "Next"
Fill in parameters as below example WITH YOUR OWN ID’s and details
- Stack-name: smart-mirror-parameters (descriptive name for stack)
- AlexaGadgetSecret: One you created in Alexa developer console
- AmazonId: One you created in Alexa developer console
- DeviceName: Hostname you gave to your raspberry Pi (recommendation use smart-mirror as hostname - it is also the default)
- LedCount: LED Count in your strip (should be 60)
- LedGPIO: GPIO Pin used in your RaspBerry Pi (21 for Raspberry Pi 4)
- LedType: Type of your Strip
Enter Stack Name "SmartMirrorDeviceRole" and press "Next"
On next screen leave defaults and press "Next"
In Final screen Acknowledge Changes if asked and select "Create Stack"
Log in with SSH to your Raspberry Pi (as pi user)
Run following commands to install amazon-ssm-agent - include quotation marks in command as below for Activation Code and ID
NOTE: use Activation Code and Activation ID you created in SSM Hybrid Activation
NOTE: Use region where you deployed your CodeStar - in example below we use eu-west-1 (Ireland)
wget https://s3.eu-west-1.amazonaws.com/amazon-ssm-eu-west-1/latest/debian_arm/amazon-ssm-agent.deb
sudo dpkg -i amazon-ssm-agent.deb
sudo service amazon-ssm-agent stop
sudo amazon-ssm-agent -register -code "Activation Code" -id "Activation ID" -region "eu-west-1"
sudo service amazon-ssm-agent start
Log in to AWS account as Admin user and open Systems Manager - Fleet Manager
NOTE: Make sure region is one you deployed - we use Ireland (eu-west-1) in our example
Select instance on above list and in Tags -Tab - Add Tags as below. Remember to "Save".
NOTE: Tags are case sensitive!
-
Name: smart-mirror (hostname of your raspberryPi device)
-
project: smart-mirror
Select "Save"
This will trigger automatic association and deploys dependencies and code deploy agent in your Raspberry Pi
After a while (up to 5 minutes) Raspberry PI should have association in place
You can check the status in the SSM / Instance / Associations:
Other place to see progress is in State Manager (In AWS Systems Manager)
After this is successful you see Association Status “Success”:
Log in to AWS account as Admin user - make sure region is where you deployed. We use Ireland (eu-west-1)
Open AWS IAM Service and add following policies to CodeStarWorker-smart-mirror-CodeDeploy Role
Add AWSCodeDeployRole and AWSCodeDeployFullAccess policies to "CodeStarWorker-smart-mirror-CodeDeploy" role. Select "Attach policies" and add policies as in picture below
Browse to Applications - Create Application
Fill in as below and press "Create Application"
Next Select "Create Deployment Group"
Fill in details as below screenshots
Add Tag Group project smart-mirror
Note Remove "Enable Load Balancing" selector
After inputting values select “Create Deployment Group”
Open AWS CodeStar smart-mirror project and edit the smart-mirror-pipeline (Through AWS Codepipeline or CodeStar service):
Press “Edit”
Select “+ Add stage” after "Source"
Stage name "DeployToDevice"
Select "Add Stage"
Select "+ Add Action Group" in DeployToDevice stage you created
Action Name: DeployToDevice
Other values as in below screenshot (Region value one you used to deploy!)
Press “Done“
"Save" your changes
Finally press “Release change” to deploy to your Raspberry Pi
This starts deployment
See additional information in https://github.com/alexa-samples/Alexa-Gadgets-Raspberry-Pi-Samples#installation
Log in to your Raspberry Pi as pi user.
NOTE: You can ignore errors like below:
tools/rctest.c: In function ‘recv_mode’:
tools/rctest.c:507:19: error: ‘SIOCGSTAMP’ undeclared (first use in this function); did you mean ‘SIOCGRARP’?
if (ioctl(sk, SIOCGSTAMP, &tv) < 0) {`
Run following commands as pi user:
sudo systemctl stop smart-mirror
cd /home/smart-mirror/script
sudo python3 setup_gadget.py --setup
This will setup the Alexa Gadget side in your Raspberry Pi
“AGREE” with the terms when prompted
> The Alexa Gadgets Raspberry Pi launch script provided herein will retrieve the 'Bluez-5.50' package at install-time from third-party sources. There are terms and conditions that you need to agree to abide by if you choose to install the 'Bluez-5.50' package (https://git.kernel.org/pub/scm/bluetooth/bluez.git/tree/COPYING?h=5.50). This script will also enable you to modify and install the 'bluez-5.50' package to enable notification callbacks after reconnections to a paired Echo device. This is required for communication between your gadget and the Echo device over BLE. If you do not agree with every term and condition associated with 'Bluez-5.50', enter 'QUIT', else enter '**AGREE**'.
Type AGREE
“Which transport mode would you like to configure your gadget for (ble/bt)?”
Type bt
Restart smart-mirror service after setup is finished
sudo systemctl start smart-mirror
Device is now in pairing more → Go and pair your Alexa Gadget with your Echo device (Bluetooth).
See additional details in Pair and Connect a Gadget to an Echo Device Over Classic Bluetooth
DONE! Happy Mirroring - you now have MVP product installed and you can continue to add functionality through challenges!