When resizing or changing a database instance such as RDS, it is important to know that a Terraform Enterprise (TFE) restart would also be needed to ensure proper functionality. Failure to do so can lead to a variety of errors that can disrupt normal TFE operations.
Here are three error messages that may be encountered if a TFE instance is not restarted after resizing or changing a database instance:
Timed out while fetching the configuration version
This error occurs when running Terraform plans/applies. It indicates that TFE is unable to fetch the configuration version within the specified time limit.
msg=SettingStorage::Postgres failed to look up setting 'atlas.base_url'
This error message indicates a generic database problem that is logged in the container
All of these error messages can be resolved by completely stopping and restarting Terraform Enterprise
If these errors persist after restarting Terraform Enterprise, please contact HashiCorp support for further assistance, and we'll be glad to assist.