LowCodeStacks
Sign in

Move a cloud flow into a solution, and fix 'missing dependencies' on import

Why flows belong in solutions, how to add existing flows (one at a time or in bulk), what to set up so they deploy cleanly, and how to read and fix 'Import failed due to missing dependencies'.

Power AutomateFix · Tutorial5 min read

Checked against Microsoft Learn · 10 sourcesHow we write guides

A flow built straight in My flows works fine until you need to move it to test or production, hand it to someone else, or call it from an agent. Then you need it in a solution: a package of the flow and everything it depends on, which is how Power Platform moves work between environments. This guide covers getting flows in, and getting solutions to import.

Why bother

Flows in solutions ("solution-aware" flows) can:

  • move between environments with their dependencies, by exporting and importing the solution, or through pipelines;
  • use connection references, so each environment plugs in its own connections at import time;
  • use environment variables for values that change per environment, such as a site URL or an email address, instead of hard-coding them;
  • change owner in place. A flow outside a solution can't, because its owner is part of its identity;
  • be used by Copilot Studio agents, which need flows to be in a solution.

Tip

New environments with Dataverse create new cloud flows inside solutions by default. Admins can check this under Settings → Product → Features → Create new canvas apps and cloud flows in Dataverse solutions.

Add an existing flow to a solution

  1. In Power Automate, go to Solutions and open your solution, or create one with your own publisher.
  2. Select Add existing → Automation → Cloud flow.
  3. Choose the Outside Dataverse tab for flows not yet in any solution. Flows already in other solutions are on From Dataverse.
  4. Select the flow and select Add.

To move many flows at once, an admin can run the PowerShell cmdlet Add-AdminFlowsToSolution to move some or all flows outside solutions in an environment.

Some flows outside solutions can't be added. Microsoft lists these cases under the solution's known limitations.

Make it deploy cleanly

Before the first export:

  • Connection references:
    • check the solution's Connection References and that each flow action uses one;
    • on import, you'll pick or create a connection for each reference in the target environment;
    • an unmapped reference makes actions fail with ConnectionNotConfigured.
  • Environment variables: replace hard-coded values such as site URLs, list names, recipients and IDs with environment variables, and set each environment's value at import time.
  • Child flows: add any child flow the parent calls to the same solution.
  • Only what you use: add only the parts of a table you actually changed (specific columns, views, forms), not the whole table. Whole tables drag in extra dependencies and layers.

"Import failed due to missing dependencies"

The solution refers to something that exists in the source environment but not in the target. Select Show dependencies. The page groups the missing items into three sections.

SectionIt meansFix
ApplicationsA Dynamics 365 or Microsoft app is missing or older in the targetInstall or update it (an admin can use the Install / Update buttons). For first-party apps, the Deploy Dependencies option can install them and then import
Managed SolutionsYour solution builds on another managed solution the target doesn't haveImport the same version of that solution into the target first
Unmanaged ComponentsSomething you customized in the source but didn't include, such as a column, view or formGo back to the source, add those components to your solution, export again, and import

Tip

Not sure what's missing? Unzip the solution file and open solution.xml. The <MissingDependencies> element lists each required component by name and ID.

Other import errors and warnings:

  • "There was an error calculating dependencies … Missing component id [GUID]": search solution.xml for that GUID to find the component's name, then add it to the solution or to the target.
  • Version warning ("exported from an environment with a more recent version of Dataverse"): the target is behind the source. You can continue, but missing components are likely, so keep development, test and production on the same versions.
  • A managed solution can only depend on managed components. If you import as managed, everything it needs must also be managed in the target.

Habits that prevent it

  • Build in a development environment, in an unmanaged solution, and deploy to test and production as managed solutions.
  • Don't edit managed components directly in test or production. That creates an unmanaged layer that hides future updates.
  • Keep environments on the same versions, and install standard first-party solutions everywhere before importing ones that depend on them.
  • Avoid dependencies on deprecated apps. They can't be installed in a new environment.

Sources