Amazon Server Setup Guide

NOTE: This page is under construction – the setup process outlined below will be tested in August 2022

By default, MindLogger response data are end-to-end encrypted and stored on the non-profit Child Mind Institute’s AWS account. We provide instructions below for owners of MindLogger applets who need to have all response data from their applets stored on their own AWS account’s server. If this situation applies to you, please contact [email protected] before continuing with the instructions below.

Information MindLogger developers will need to set up your AWS server connection
Set up AWS server

Information MindLogger developers will need to set up your server connection

Routing encrypted data collected from the MindLogger app to your AWS server requires initial setup assistance from MindLogger developers. If you follow the instructions in this guide, you will be able to provide them with the following required information to connect your server:

AWS variables to access your MindLogger Docker container:
- MindLogger account owner email address
- Mongodb database name (if changed from “mindlogger” in steps 34 and 35)
- Database root user name (steps 34, 35, 36)
- Database root password (steps 34, 35, 36)
- Database port (if different than the default 27017)
- Public IP address for your MindLogger EC2 instance (step 27)
AWS variables to access your MindLogger S3 storage bucket (for media files):
- Storage bucket name (if changed from “my-mindlogger-bucket” in steps 11 and 16)
- Access key (step 8)
- Secret access key (step 8)

Set up AWS server

Create AWS user

You do not want to share information about a user with root access to your AWS account to MindLogger developers or anyone else! The following instructions will guide you to set up a new AWS user with restricted access to a new s3 bucket for (optional) media files collected by your MindLogger applets.

Sign in to https://aws.amazon.com/:
Enter “IAM” in the search box at top and select “Users” under “IAM”:
In the IAM > Users screen, tap the blue “Add users” button:
In the “Add user” screen #1:
- Enter “mindlogger-user” as the User name
- Select both credential types (Access key and Password)
- Tap the blue “Next: Permissions” button at the bottom
In the “Add user” screen #2, tap the blue “Next: Tags” button:
In the “Add user” screen #3, tap the blue “Next: Review” button:
In the “Add user” screen #4, tap the blue “Create user” button:

In the “Add user” screen #5 (screenshot below), download the .csv table
*** SHARE THIS INFORMATION ONLY WITH MINDLOGGER ADMINISTRATORS ***
[see top]

The file contains:
- the Access key ID for the new user
- the Secret access key for the new user
Tap the “Close” button
You will then be taken to the IAM > Users screen with the new mindlogger-user
Tap “mindlogger-user”
Tap the “Add inline policy” link on the right:

Copy/paste the text below into the “Create policy” page “JSON” tab (ref [1] and [2])

Tap the blue “Review policy” button:

{
    "Version": "2012-10-17",
    "Statement": [
        {
            "Effect": "Allow",
            "Action": [
                "s3:ListBucket"
            ],
            "Resource": [
                "arn:aws:s3:::my-mindlogger-bucket"
            ]
        },
        {
            "Effect": "Allow",
            "Action": [
                "s3:PutObject",
                "s3:GetObject",
                "s3:DeleteObject"
            ],
            "Resource": [
                "arn:aws:s3:::my-mindlogger-bucket/*"
            ]
        }
    ]
}

Enter “mindlogger-user-policy” in the Name field
Tap the blue “Create policy” button:
You will then be taken to the IAM > Users screen with the new mindlogger-user and mindlogger-user-policy:

Create AWS S3 bucket

This section will guide you to create a new AWS s3 bucket for (optional) media files collected by your MindLogger applets.

Enter “s3” in the search box at top and select “Buckets” under “S3”:
In the Buckets page, tap the orange “Create bucket” button:

Enter “my-mindlogger-bucket” into the Bucket name field
Tap the orange “Create bucket” button at the bottom (not visible in the screenshot):
You can now see the my-mindlogger-bucket in the list of S3 buckets:

Create AWS EC2 instance

Your MindLogger database needs to be set up somewhere to store the response data collected by your MindLogger applets. This section will guide you to create a new AWS Linux 2 instance.

Enter “ec2” in the search box at top and select “Instances” under “EC2”:
In the Instances page, tap the orange “Launch instances” button:
Enter “mindlogger-instance” in the Name field
If you tap the orange “Launch instance” button…
…it will alert you that you need a Key pair name
Tap on the “Create new key pair” link:
Enter “mindlogger-key-pair” in the Name field
Tap the orange “Create key pair” button:
Tap the orange “Launch instance” button:
It should then confirm you have successfully initiated launch of the new instance
Tap the orange “View all instances” button:
You should see mindlogger-instance in the list of EC2 instances (you might need to wait and refresh the Instances page):

Create MindLogger database in EC2 instance

This section will guide you to create an empty MindLogger mongodb database. You will copy a Docker container with a mongodb database from a public AWS Elastic Container Registry repository to your new AWS EC2 instance.

Tap Instance ID link for mindlogger-instance in the Instances page just above
Tap the “Connect” button in the mindlogger-instance page:
Select the “EC2 Instance Connect” tab in the “Connect to instance” page
Tap the orange “Connect” button:
In the EC2 Instance Connect screen, type the command sudo yum update:
Whenever “[y/d/N]” is presented, choose “y” for “yes”:
Type the command sudo yum install docker:
Enable docker service at AMI boot time by typing the command:

sudo systemctl enable docker.service
Start the Docker service with the command:

sudo systemctl start docker.service

Pull the public MindLogger container image from Amazon’s Elastic Container Registry:

sudo docker pull public.ecr.aws/c4l1a6f8/mindlogger:latest
- (See Appendix for how the public MindLogger container was built)
Run the image as a mindlogger container by typing the command:

sudo docker run -it -d --name mindlogger public.ecr.aws/c4l1a6f8/mindlogger

Reset database username and password

For security, you must change at least the password that accesses your database. This section will guide you to create a new user with access to your mongodb database and change relevant environment variables. You will need to share this information with MindLogger administrators.

Set environment variables (do not change once connected)
*** SHARE THIS INFORMATION ONLY WITH MINDLOGGER ADMINISTRATORS ***
[see top]

NOTE: replace new_username and new_password with new login credentials

echo "export MONGO_INITDB_DATABASE=mindlogger" >> .bashrc

echo "export MONGO_INITDB_ROOT_USERNAME=new_username" >> .bashrc

echo "export MONGO_INITDB_ROOT_PASSWORD=new_password" >> .bashrc

source .bashrc

Reset the database username and password inside the mongodb Docker container:

NOTE: replace new_username and new_password with the new login credentials

sudo docker exec -it mindlogger bash -c mongo \
--username new_username --password new_password

At the command line prompt (>), create a new user with the new login credentials:

NOTE: replace new_username and new_password with the new login credentials

db.createUser({user: 'new_username', pwd: 'new_password', \
roles: [{ role: 'readWrite', db:'[database]'}]})
Type Ctrl-C to escape this prompt and exit to exit the console

Appendix: How the public MindLogger Docker container was built

The information below is for completeness and transparency, and gives the steps used to create the AWS ECR public MindLogger container that is used in the instructions above.

Build the Docker container in an EC2 instance on the same account that owns the ECR repository:
- sudo yum update
- sudo yum install git
- sudo yum install docker
- sudo curl -L "https://github.com/docker/compose/releases/download/1.24.1/docker-compose-$(uname -s)-$(uname -m)" -o /usr/local/bin/docker-compose
- sudo chmod +x /usr/local/bin/docker-compose
- sudo systemctl enable docker.service
- sudo systemctl start docker.service
- git clone https://github.com/ChildMindInstitute/mindlogger-arbitrary-server.git
- cd mindlogger-arbitrary-server
- sudo /usr/local/bin/docker-compose build
- sudo /usr/local/bin/docker-compose up -d
Set up AWS CLI v2:
- curl "https://awscli.amazonaws.com/awscli-exe-linux-x86_64.zip" -o "awscliv2.zip"
- unzip awscliv2.zip
- sudo ./aws/install
- /usr/local/aws-cli/v2/current/bin/aws configure
Set up a public repo on AWS’s ECR ([1], [2], [3]):
- Attach AmazonElasticContainerRegistryPublicFullAccess policy to the user
- aws ecr-public get-login-password --region us-east-1 | sudo docker login --username AWS --password-stdin public.ecr.aws/c4l1a6f8
- Push the Docker container to the public repo on ECR:
  - sudo docker tag mongo:4-bionic public.ecr.aws/c4l1a6f8/mindlogger:latest
  - sudo docker push public.ecr.aws/c4l1a6f8/mindlogger:latest