dYdX Chain Infrastructure

This repository contains Terraform related configs for dYdX chain used for testing purposes.

Terraform Cloud

We use Terraform Cloud to deploy and manage infrastructure resources. Terraform Cloud helps with Terraform state management as well as automating the deployment pipeline (i.e. deploying infra changes made in this Github repo).

Install the terraform cli using tfenv. We currently use version 1.3.2 of terraform.

Terraform Cloud account for dYdX v4: link

How to test local infra changes against "dev" env

This workspace is configured using The CLI-driven Run Workflow.

1. Install terraform if you haven't done so already following the instructions here.
   1. Running terraform --version should output a version >= v1.3.2.
2. Login to Terraform Cloud via terraform login. You will be asked to create a token and provide it to the CLI.
3. cd into the relevant folder, for example /indexer, /metrics_ingestor, or /indexer_dashboards.
4. Initialize terraform if you haven't done so already with terraform init.
5. Run terraform workspace select indexer-dev to select the dev workspace for validators. You can also run terraform workspace list to see available workspaces.
6. Once you've selected the indexer-dev workspace, feel free to run terraform plan and terraform apply to trigger your runs. Note that these runs will still take place remotely in Terraform Cloud, but the output will be streamed to your machine.

How to apply changes to infra to "staging" env

1. Go to the corresponding Terraform Cloud workspace (i.e. https://app.terraform.io/app/dydxprotocol/workspaces/validators-staging)

2. Go to Runs tab

3. Identify or Trigger Run:

• If you recently merged a PR to the main branch, then you would see a Run already triggered. Use the triggered Run in this case.
• Otherwise, trigger a new Run by selecting Actions > Start new run. Use the newly triggered Run in this case.

Note: in some cases, if previous runs have not been applied, you will see a list of runs that are queued. In this case, one way to resolve this is to cancel all previous runs and then triggering a new fresh run via UI.

4. Review the plan output. Ensure that the resources being created, updated or destroyed are intended.

5. Confirm & Apply via UI.

Datadog

For datadog-specific instructions, follow the instructions here