How to Set Up S3 Buckets, AWS Users, and Heroku ENV Variables for Uploading to Coyote

This page walks through how to create AWS IAM Custom Policies, Groups, and Users and S3 Buckets. Then how to update those environmental variable in Heroku.

IN AWS:

Create Custom Policies:

• Access IAM policies
• Create a policy
• Select to view JSON editor
• Enter this code:

{
    "Version": "2012-10-17",
    "Statement": [
        {
            "Effect": "Allow",
            "Action": "s3:ListAllMyBuckets",
            "Resource": "arn:aws:s3:::*"
        },
        {
            "Effect": "Allow",
            "Action": "s3:*",
            "Resource": [
                "arn:aws:s3:::production.coyote",
                "arn:aws:s3:::production.coyote/*"
            ]
        }
    ]
}

• Name: production.coyote / staging.coyote
• Description: This group is allowed full access to the production.coyote S3 bucket only.

Create IAM Groups:

• Access IAM Groups
• Create New Group
• Name: production.coyote
• Apply production.coyote policy to new group

Create Users: Create separately so you can apply separate groups

• Add Users
• User name: production.coyote / staging.coyote
• Access Type: Programmatic access
• Add user to group: production.coyote / staging.coyote
• Download the access key .csv so you can update the env variables on Heroku

---

Update Heroku:

• From your terminal, run heroku config --app stagingcoyote (and the same for livecoyote) to see what environmental variables need updating
• You will see that you need to update the following environmental variables with new keys:
  • S3_ACCESS_KEY:
  • S3_BUCKET:
  • S3_SECRET_KEY:
• Update the variables with the following commands (do for staging and production:
  • heroku config:set S3_ACCESS_KEY=theaccesskey --remote staging
  • heroku config:set S3_BUCKET=thebucketname --remote staging
  • heroku config:set S3_SECRET_KEY=thesecretkey --remote staging
• Double check by running config again to see that everything has updated properly

---

• Test staging by logging in, uploading a photo, then check the corresponding S3 bucket in AWS to ensure the file appears.