Just noticed that the genesyscloud_edge_configuration resource is failing during the plan phase with a state drift error, even though the configuration has not been modified in the local state file. The deployment pipeline, running on GitHub Actions in apac-1, reports that the analytics_aggregation block has been changed outside of Terraform.
Environment details:
- Genesys Cloud Terraform Provider: v1.15.2
- Genesys Cloud CLI: v2.2.0
- Region:
apac-1 - Infrastructure: BYOC Edge deployment
The specific error output is as follows:
Error: Configuration has changed outside of Terraform
The plan failed to produce any changes because the configuration has changed outside of Terraform since the last apply. The following attributes have been modified:
~ analytics_aggregation {
~ retention_days = 30 -> 90
}
The local HCL defines retention_days as 30. However, checking the Genesys Cloud admin portal confirms the setting is indeed 30. When I run terraform refresh, the state file updates to reflect 90, which is incorrect. Subsequent terraform plan commands then show no drift, but the state is now out of sync with the actual API response from the Edge node.
This issue is critical because it breaks our automated compliance reporting. The analytics data is not being aggregated correctly for the BYOC edge nodes, leading to inconsistent metrics in the Power BI dashboards. We rely on the Terraform state to be the source of truth for our CX-as-Code strategy. If the provider cannot accurately reflect the configuration on the Edge, we cannot trust the deployment pipeline.
Has anyone encountered this specific behavior with BYOC Edge analytics settings? Is this a known bug in the provider version 1.15.2? We are considering reverting to a previous provider version, but that introduces other risks with newer feature support. Any insights on how to force the provider to read the correct value from the Edge API would be appreciated. We have already cleared the cache and re-authenticated the service account, but the issue persists.