Open SensitiveSKIN opened 3 years ago
Marking this issue as stale due to inactivity. This helps our maintainers find and focus on the active issues. If this issue receives no comments in the next 30 days it will automatically be closed. Maintainers can also remove the stale label.
If this issue was automatically closed and you feel this issue should be reopened, we encourage creating a new issue linking back to this one for added context. Thank you!
Not stale.
Community Note
Terraform CLI and Terraform AWS Provider Version
Affected Resource(s)
aws_vpc_endpoint
Terraform Configuration Files
Expected Behavior
Terraform should wait for the GWLB-Endpoint to reach the "available" state (which it will, after a minute or two) before moving on and/or failing the complete.
Actual Behavior
Truly, the relevant portion of the configuration above is the GWLB-Endpoint. This error only seems to occur when the GWLB-Endpoint is set to "auto_accept = true". If this is set to false, Terraform will complete the process just fine. Strangely, even when "auto_accept" is set to true and Terraform fails, the GWLB-Endpoints will actually create, but the current versions of the AWS provider apparently expect it to be nearly instant, when in reality, it can take at least a minute or two for the GWLB-Endpoints to create, then have their connection to the Endpoint Service accepted. It should be noted that the timeouts above do nothing to improve (or degrade) the performance issues.
Steps to Reproduce
Important Factoids
This functionality worked in AWS Provider version 3.38.0. In fact, if I revert to 3.38.0, I can still create the GWLB-Endpoints on the first try. I believe it was initially fixed in aws_vpc_endpoint does not autoaccept #14604, but the branch which contained the issue must have been merged back into the main branch, as it does not work in 3.53.0 (or many of the recent versions).
References