#atmos - June 2025

I changed my atmos config from name_pattern to name_template, but it doesn't seem to work with the Atlantis Integration yet?

 Error

stack name pattern must be provided in 'stacks.name_pattern' config or 'ATMOS_STACKS_NAME_PATTERN' ENV variable

I used to be able to write !terraform.output component stack yq_expression["index"] but now get FATA parse error on line 1, column 68: bare " in non-quoted-field after upgrading to v1.179.0. I've tried a variety of ways of escaping but none seem to work.

@Igor Rodionov can you please delete these release announcements when you are testing?

I am brand new to Atmos and I'm trying to customize my backend state storage key. I thought to use templates to do this, but it doesn't seem to be working. Am I doing something wrong? Or misunderstanding how templates work?

staging.yml

terraform:
  backend_type: azurerm
  backend:
    azurerm:
      storage_account_name: foo
      subscription_id: foo
      tenant_id: foo
      container_name: tfstate
      use_azuread_auth: true
      resource_group_name: foo
      key: "{{ .vars.client_code }}/{{ .vars.client_code }}-{{ .vars.product_code }}-{{ .vars.environment }}-{{ .vars.location_code }}"

This file is imported from the specific client folder in _defaults.yaml (I'm following roughly the "component inheritance" design pattern):

import:
  - mixins/tenant/staging

However, end result via atmos describe component is that the values don't get replaced.

        key:
            final_value: '{{ .vars.client_code }}/{{ .vars.client_code }}-{{ .vars.product_code }}-{{ .vars.environment }}-{{ .vars.location_code }}/customer-init.terraform.tfstate'
            name: key
            stack_dependencies:
                - dependency_type: import
                  stack_file: mixins/tenant/staging
                  stack_file_section: terraform.backend.azurerm
                  variable_value: '{{ .vars.client_code }}/{{ .vars.client_code }}-{{ .vars.product_code }}-{{ .vars.environment }}-{{ .vars.location_code }}'

I know that the .vars.* I'm referencing exists in atmos because in my atmos.yaml I have the following that does get replaced correctly:

  name_template: "{{.vars.client_code}}-{{.vars.product_code}}-{{.vars.environment}}-{{.vars.location_code}}"

Edit: to clarify, I'm trying to accomplish something similar to what we previously did with key = "${path_relative_to_include()}/terraform.tfstate" in Terragrunt.

Hey all, getting used to atmos but trying to get over a major hurdle, I want to be able to define one file that is reused by multiple stacks, and I was assuming I could use the !env yaml function to do this, I have a pretty simple configuration:

import:
  - deploy/aws/customer/test/_defaults
  - mixins/customer/_us-customer

vars:
  environment: us
  customer_id: !env CUSTOMER_ID

But I get this error (slightly redacted):

Could not find the component 'customer_test' in the stack 'aws_us_ABC123'.

when running

CUSTOMER_ID=ABC123 atmos tf plan customer_test -s aws_us_$CUSTOMER_ID

If I run

CUSTOMER_ID=ABC123 atmos describe stacks -s aws_us_$CUSTOMER_ID

I just get a blank object {}

I have narrowed this down to the use of the !env function, since if I replace the !env function with hard coded values, it works fine.

I have been trying to debug it but not getting a ton of helpful information. Any help is appreciated! thanks

Hi all,

I am trying to run some workflows with atmos. However, I am getting this error:

Error: Failed to load plugin schemas
│ 
│ Error while loading schemas for plugin components: 7 problems:
│ 
│ - Failed to obtain provider schema: Could not load the schema for provider registry.opentofu.org/cloudposse/awsutils: failed to instantiate provider "registry.opentofu.org/cloudposse/awsutils" to obtain schema: unavailable provider "registry.opentofu.org/cloudposse/awsutils".
│ - Failed to obtain provider schema: Could not load the schema for provider registry.opentofu.org/cloudposse/utils: failed to instantiate provider "registry.opentofu.org/cloudposse/utils" to obtain schema: unavailable provider "registry.opentofu.org/cloudposse/utils".
│ - Failed to obtain provider schema: Could not load the schema for provider registry.opentofu.org/hashicorp/aws: failed to instantiate provider "registry.opentofu.org/hashicorp/aws" to obtain schema: unavailable provider "registry.opentofu.org/hashicorp/aws".
│ - Failed to obtain provider schema: Could not load the schema for provider registry.opentofu.org/hashicorp/helm: failed to instantiate provider "registry.opentofu.org/hashicorp/helm" to obtain schema: unavailable provider "registry.opentofu.org/hashicorp/helm".
│ - Failed to obtain provider schema: Could not load the schema for provider registry.opentofu.org/hashicorp/kubernetes: failed to instantiate provider "registry.opentofu.org/hashicorp/kubernetes" to obtain schema: unavailable provider "registry.opentofu.org/hashicorp/kubernetes".
│ - Failed to obtain provider schema: Could not load the schema for provider registry.opentofu.org/hashicorp/random: failed to instantiate provider "registry.opentofu.org/hashicorp/random" to obtain schema: unavailable provider "registry.opentofu.org/hashicorp/random".
│ - Failed to obtain provider schema: Could not load the schema for provider registry.opentofu.org/integrations/github: failed to instantiate provider "registry.opentofu.org/integrations/github" to obtain schema: unavailable provider "registry.opentofu.org/integrations/github"..

the commands that I am running are:

atmos terraform destroy eks/argocd -s plat-usw2-prod

atmos workflow delete/argocd --file argocd.yaml 

argocd.yaml

[...]
  delete/argocd:
    description: |
      This workflow deletes ArgoCD components in the correct order
    steps:
      - command: terraform destroy eks/argocd -s plat-usw2-prod
      - command: terraform destroy eks/argocd -s plat-usw2-staging
      - command: terraform destroy eks/argocd -s plat-usw2-dev

  delete/argocd-repos:
    description: |
      This workflow deletes ArgoCD repos
    steps:
      - command: terraform destroy argocd-deploy-prod -s core-gbl-auto
      - command: terraform destroy argocd-deploy-non-prod -s core-gbl-auto

Sorry if this has been asked before - is there a recommended way to pass inputs and outputs between different stacks?

I'm attempting to demo Atmos for my new company (I'm a terragrunt convert). I got a quick and dirty demo with local state working, and am now trying to move over to a remote s3 backend following this documentation. (My demo requires remote-state for passing outputs between stacks).

https://atmos.tools/quick-start/advanced/configure-terraform-backend/#terraform-s3-backend

I'm able to apply the module, following the instructions in the git repo to -force-copy etc after the s3 and dynamo are up. After I create the default s3 backend, whenever I try to apply another stack I get this error:

missing state name

Any ideas to get unblocked? This is probably just a silly omission on my end but I can't pin it down even though I've followed the docs closely.

I've run into
runtime: goroutine stack exceeds 1000000000-byte limit
fatal error: stack overflow
issue during atmos apply for the first time. Any advice how to go around that?

There seems to be an issue with OpenTofu 1.10.0 and Atmos, related to the UserAgent postfix that Atmos adds.
Is there a workaround for this?

/github unsubscribe cloudposse/atmos releases

This channel will receive notifications from cloudposse/atmos for:

😢 having a real hard time of a time using atmos to store SSM output using the hook if the output has templating characters in it {{ }}

Trying to understand how to implement components hosted Cloud Posse Terraform Components. Do I add to the vendor manifest and download contents in src directory?