terraform remote state has to be run twice

[myoung34]

Hi there,

Thank you for opening an issue. Please note that we try to keep the Terraform issue tracker reserved for bug reports and feature requests. For general usage questions, please see: https://www.terraform.io/community.html.

Terraform Version

0.7.5

Affected Resource(s)

1. terraform_remote_state

Technically affects all resources

Terraform Configuration Files

  terraform remote config \
    -backend=s3 \
    -backend-config="bucket=${BUCKET}" \
    -backend-config="key=${LOCATION}/terraform.tfstate" \
    -backend-config="region=us-east-1"

Debug Output

https://gist.github.com/myoung34/c2343ed930e25899718c5c14cea7290b

Expected Behavior

Should have succeeded in initializing remote states

Actual Behavior

Failed and bombed out on first initialize. Fixed by running bash init.sh || bash init.sh
Succeeds on second run every time

Steps to Reproduce

1. Use something like Jenkins EC2 plugin which runs outside of 'userdata'
2. Run terraform remote config -backend=s3 near immediate start of instance

Important Factoids

Running from Jenkins EC2 slave instance using IAM role. Will fail the first time ever run. All subsequent uses of terraform remote config will succeed.

References

#5839 - see @sjpalf 's comment

[ekulz]

This is also happening to me. Remote state would pull down fine but on apply it kept giving 403 errors. My steps were (using a python wrapper):

terraform.py remote
terraform.py get
terraform.py plan
terraform.py apply

Once I added another remote step in between plan and apply it worked perfect? On version 0.7.7

Nevermind, all looks ok now. Think it was a glitch when copying my state file to a server side encryption required S3 bucket. Ever since adding that second remote step and performing a proper apply, I have been able to remove the second remote step and everything runs fine.

[apparentlymart]

Hi @myoung34 and @lhickin! Sorry for the issues here and sorry we let this issue languish.

The remote system was completely redesigned for 0.9, so the workflow this issue is discussing is no longer present. Given how much the relevant parts of Terraform have changed since 0.7.5 I'm going to close this. If anyone is seeing similar issues in 0.9 or above I suggest opening a fresh issue so that we can triage and address it separately. Thanks for opening this issue!

[ghost]

I'm going to lock this issue because it has been closed for 30 days ⏳. This helps our maintainers find and focus on the active issues.

If you have found a problem that seems similar to this, please open a new issue and complete the issue template so we can capture all the details necessary to investigate further.