Geoservices for AWS

Amazon CDK configurations for deploying Princeton University Library geoservices.

First-time setup

• Install language dependencies with asdf or according to versions listed in .tool-versions

• Install aws-cdk client

  brew install aws-cdk
  

• Install pipenv

  pip install --user pipenv
  

• Install the required python dependencies

  pipenv sync
  

• Activate your virtualenv

  pipenv shell
  

• Update your .aws/config to include:

  [geoservices-deploy]
  region = us-east-1
  

• Update your .aws/credentials to include credentials from LastPass -> Shared-ITIMS-Passwords\Figgy -> TiTilerAWS like

  [geoservices-deploy]
  aws_access_key_id = [username]
  aws_secret_access_key = [password]
  

• Copy .env.example to .env and update the account number using the note from that LastPass entry.

Every time setup

pipenv sync
pipenv shell

Check that changes are valid

cdk synth

Deploy Geodata

1. Deploy the staging stack

cdk --profile geoservices-deploy deploy geodata-staging

1. Deploy the production stack

cdk --profile geoservices-deploy deploy geodata-production

Deploy TiTiler

1. Deploy the staging stack

cdk --profile geoservices-deploy deploy titiler-staging

1. Deploy the production stack

cdk --profile geoservices-deploy deploy titiler-production

To add additional dependencies, for example other CDK libraries, just add them with the pipenv install command.

Useful commands

Read the CDK documentation

• cdk deploy deploy this stack to your default AWS account/region
• cdk destroy delete this stack from your default AWS account/region
• cdk diff compare deployed stack with current state
• cdk docs open CDK documentation
• cdk ls list all stacks in the app
• cdk synth emits the synthesized CloudFormation template