search

mrolla / terraform-provider-circleci

Terraform provider for CircleCI
MIT License
96 stars 37 forks source link

Plans applied through mac M1 are not recognised as applied by mac running on intel CPUs #141

Closed atmosx closed 9 months ago

atmosx commented 1 year ago

Hello,

I'm using this plugin using two mac laptops. I'm working mainly on a macbook pro M1 with ARM64 CPU, however every now and then I have to do some work from an older macbook pro running on Intel (x86) CPU.

When I create, plan and apply resources using the M1 laptop the state looks fine. Now on the intel laptop, running terraform plan will show the circleCI resources - only the circleCI resources - as new and will try to re-create them. Of course this will fail, as the env variables exists already.

This is a bit of a problem as some team members have mac with x86 CPUs and others have Macs with ARM CPUs. I have not tested what happens when running on linux but my guess is that this is a CPU specific more than OS specific issue.

Kind regards,

P.

kelvintaywl commented 1 year ago

hi there, would you know where your state is saved in?

in other words, is your Terraform state saved locally or via a remote backend (e.g., AWS S3)? https://developer.hashicorp.com/terraform/language/settings/backends/configuration

If local, this is then not an issue with this Terraform provider (or any other provider); local backends (default) mean the state is saved as a file on your local machine, and switching to other machines will see the issue you mentioned above 🤓 hope it helps!

atmosx commented 9 months ago

hi there, would you know where your state is saved in?

in other words, is your Terraform state saved locally or via a remote backend (e.g., AWS S3)?

https://developer.hashicorp.com/terraform/language/settings/backends/configuration

If local, this is then not an issue with this Terraform provider (or any other provider);

local backends (default) mean the state is saved as a file on your local machine, and switching to other machines will see the issue you mentioned above 🤓 hope it helps!

Sorry for the late reply here. This is not a problem anymore for me.

The state was/is remote, stored in s3.

kelvintaywl commented 9 months ago

@atmosx thanks for the updates! perhaps you can close this issue too then :)