Upgraded our dev and prod Atlantis systems from

v0.20.1

to `v0.23.5`with zero problems yesterday!

I’m glad to hear that, we had a few regresions and other issues so I’m glad is working for you

We use atlantis to deploy all of our terraform changes. Environment: 100% AWS, but also Snowflake, Databricks, Github, Pagerduty, etc. managed via terraform. Atlantis running in ECS Fargate, Elasticache Redis for locking, EFS for plan storage (it literally deploys itself to ECS. I am writing a blog about this soon-ish) repos.yaml:

repos:
  - id: /.*/
    # allowed_overrides specifies which keys can be overridden by this repo in
    # its atlantis.yaml file.
    allowed_overrides: [apply_requirements, workflow]
    apply_requirements: [approved,mergeable]
    # allow_custom_workflows defines whether this repo can define its own
    # workflows. If false (default), the repo can only use server-side defined
    # workflows.
    allow_custom_workflows: true
    # This hook is to work around <https://github.com/runatlantis/atlantis/issues/2221>
    pre_workflow_hooks:
      - run: git config --global --add safe.directory '*'


# Defining default and release workflows here to centralize simplify for self-service repos
workflows:
  default:
    plan:
      steps:
        - init
        - plan
    apply:
      steps:
        - apply
  release:
    plan:
      steps:
        - init
        - plan:
            extra_args: ["-var-file", "environment.tfvars"]
    apply:
      steps:
        - apply

atlantis.yaml:

version: 3

parallel_plan: true
parallel_apply: true

projects:
  - dir: terraform/environments/dev
    workflow: release
    autoplan:
      when_modified: ["../../modules/**/*.tf", "*.tf*"]
  - dir: terraform/environments/prod
    workflow: release
    autoplan:
      when_modified: ["../../modules/**/*.tf", "*.tf*"]

Hi Jason, with this setup, are you able to achieve horizontal scaling with for Atlantis?

Hi! No, we do not run multiple atlantis containers in ECS although it might be possible, I’m not sure.