Partial failure of provider fails to taint resource

[ryanuber]

Droplet was created, but provisioner was never run.

digitalocean_droplet.nyc1-server1: Error: Error waiting for droplet (2896483) to become ready: couldn't find resource
Error applying plan:

1 error(s) occurred:

* Error waiting for droplet (2896483) to become ready: couldn't find resource

Terraform does not automatically rollback in the face of errors.
Instead, your Terraform state file has been partially updated with
any resources that successfully completed. Please address the error
above and apply again to incrementally change your infrastructure.
tf master » terraform plan
Refreshing Terraform state prior to plan...

digitalocean_droplet.sfo1-server3: Refreshing state... (ID: 2896436)

[ryanuber]

On a subsequent terraform apply, the resource appears to be in sync, but the provisioner will not be run at this point.

[mitchellh]

Notes: it looks like we need to set the resource as tainted if there are provisioners and there was an error from the initial Apply call.

I don't think it is tainted if there are no provisioners, or its not a creation event. Terraform is meant to handle partial state, so this should work fine. But due to us throwing away connection info, this results in a broken partial state for this case.

Thoughts @armon?

[mitchellh]

Fixed in line to my comment above

[armon]

Sounds right to me