Facing an issue while using the custom workspace from default

[Jenishk56]

Community Note

• Please vote on this issue by adding a 👍 reaction to the original issue to help the community and maintainers prioritize this request. Searching for pre-existing feature requests helps us consolidate datapoints for identical requirements into a single place, thank you!
• Please do not leave "+1" or other comments that do not add relevant new information or questions, they generate extra noise for issue followers and do not help prioritize the request.
• If you are interested in working on this issue or have submitted a pull request, please leave a comment.

---

�[31m│�[0m �[0m�[1m�[31mError: �[0m�[0m�[1mBackend initialization required, please run "terraform init"�[0m
�[31m│�[0m �[0m
�[31m│�[0m �[0m�[0mReason: Initial configuration of the requested backend "gcs"
�[31m│�[0m �[0m
�[31m│�[0m �[0mThe "backend" is the interface that Terraform uses to store state,
�[31m│�[0m �[0mperform operations, etc. If this message is showing up, it means that the
�[31m│�[0m �[0mTerraform configuration you're using is using a custom configuration for
�[31m│�[0m �[0mthe Terraform backend.
�[31m│�[0m �[0m
�[31m│�[0m �[0mChanges to backend configurations require reinitialization. This allows
�[31m│�[0m �[0mTerraform to set up the new configuration, copy existing state, etc. Please
�[31m│�[0m �[0mrun
�[31m│�[0m �[0m"terraform init" with either the "-reconfigure" or "-migrate-state" flags
�[31m│�[0m �[0mto
�[31m│�[0m �[0muse the current configuration.
�[31m│�[0m �[0m
�[31m│�[0m �[0mIf the change reason above is incorrect, please verify your configuration
�[31m│�[0m �[0mhasn't changed and try again. At this point, no changes to your existing
�[31m│�[0m �[0mconfiguration or state have been made.
�[31m╵�[0m�[0m
�[0m�[0m```
This is how the atlantis.yaml configuration looks like

• name: xyz
  dir: xyz
  workflow: xyz
  workspace: xyz
  terraform_version: 0.13.7

xyz:
plan:
steps:
- run: "terraform workspace new $WORKSPACE || true; rm -rf .terraform/"=
- init:
extra_args: ["-lock=false", "-reconfigure"]
- plan:
extra_args: ["-lock=false"]
apply:
steps:
- apply:
extra_args: ["-lock=false"]

### Overview of the Issue

I was using the `default` workspace and recently started using the custom workspace, e.g. `xyz`
I have copied the state file in the gcs bucket from `default.tfstate` to `xyz.tfstate`
It works fine from the local system while using the workspace `xyz`. While using the atlantis, I see above weird output. 
However, Atlantis plan and apply works fine but I don't want to see that irrelevant output in my plan. 

I am using Terraform `0.13.7`

[ganeshk1928]

@Jenishk56 I too ran into trouble when switching from default to a new workspace. This is because atlantis is setting the TF_WORKSPACE env. There is a PR open #1578 to fix this. Meanwhile instead of running terraform workspace new $WORKSPACE || true; rm -rf .terraform/" you can unset the TF_WORKSPACE env variable in your atlantis.yalm file. I'm adding my sample atlantis.yaml file here.

version: 3
projects:
- name: project-1
  dir: .
  workspace: staging
  autoplan:
    when_modified: ["*.tf"]
    enabled: false
  apply_requirements: [mergeable, approved]
  workflow: default
workflows:
  default:
    plan:
      steps:
      - env:
          name: TF_WORKSPACE
          value:
      - init
      - plan:
          extra_args: ["-var-file", "common.tfvars"]
    apply:
      steps:
      - run: echo "Performing Terraform apply"
      - env:
          name: TF_WORKSPACE
          value:
      - apply

[jamengual]

fixed in #2006