Hello,
I am facing an issue while working with environment variables in Power Automate Desktop (PAD) using the “Power Automate Environment” connector.
When I move a solution from one environment to another, I notice that the environmentvariabledefinitionid gets hardcoded in the solution dependencies (customization.xml). As a result, during import into the target environment, the deployment fails with an error stating that the referenced environment variable does not exist.
This happens even though:
- The same environment variable (with identical schema name) is already present in the target environment
- The environment variables are deployed separately as part of another managed/unmanaged solution
Despite this, the solution containing the desktop flow still fails to import.
After investigating the Dataverse tables, I found that the PAD connector internally binds to the environmentvariabledefinitionid (GUID) rather than using the schema name. Because this GUID is environment-specific, it does not match across environments, leading to the failure.
Interestingly, when accessing the same environment variables from cloud flows, there is no issue, as they correctly resolve using the schema name.
Has anyone else encountered this behavior with PAD? Is there any recommended approach or workaround to make desktop flows with environment variables ALM-friendly across environments?
