This gets hold of the TrustGraph AWS deployment bundle. It deploys an AWS EC2 instance with an install script which does the complete TrustGraph install.
One point to note is that the AWS EC2 instance comes up very quickly, but the installation will be running in the background and takes a couple more minutes, so make sure you track the installation to completion before wondering why things aren't working.
This drops the deploy repo in trustgraph-deploy in your home directory,
but you can put it anywhere, the location isn't important.
mkdir ~/trustgraph-deploy
cd ~/trustgraph-deploy
git clone https://github.com/trustgraph-ai/aws-memgraph-workshop .
Pulumi is an infrastructure-as-code deployment tool, and this has been used to create the deployment you are about to use.
cd pulumi
npm install
pulumi login --local
That last line tells Pulumi where to store its 'state'. If you were using this for an operational deploy, you'd want to put the state somewhere that other people can get to, like an S3 bucket, but for this workshop it doesn't matter. It says to put the state on your device.
You should set the AWS profile for your environment using the profile you created earlier.
export AWS_PROFILE=workshop
Set the Pulumi config passphrase to the empty string. This encrypts the Pulumi state, and we don't need that for this workshop.
export PULUMI_CONFIG_PASSPHRASE=''
Finally, you have to initialise the Pulumi "stack" that we're going to deploy:
pulumi stack init workshop
I should mention that this deploy creates a new VPC and installs the stuff
in that VPC. If you're using an existing AWS account that you've used
for other things, you may find the IP address range we've chosen to use
conflicts with things you already have running. The address range
is in Pulumi.workshop.yaml so modify the config in that file if you
get a conflict in the deploy.
Next: Deploy TrustGraph to AWS