Skip to content
Tutorials Center
My learning

Topics

Lesson 4 of 38 ยท Getting Started

Power Automate Designer, Testing and Run History

Find your way around the flow designer, test with manual or automatic runs, read run history and outputs, resubmit runs, and troubleshoot failed flows.

Most of your time with Power Automate goes into two places: the designer, where you edit, and the run history, where you find out what actually happened. Knowing both well makes you faster than any amount of cleverness with expressions.

The designer

When you open a flow and select Edit, you see the flow as a vertical list of cards.

  • Cards are steps. Select one to expand its inputs. The three-dot menu gives Rename, Copy to my clipboard, Add a comment, Settings, Peek code and Delete.
  • + between cards inserts a step or a parallel branch.
  • The top bar has Save, Test, Flow checker, and (in the new designer) the Copilot pane where you describe changes in plain language.
  • Settings on a card control timeouts, retry policy, pagination and concurrency.

Peek code shows the JSON of a card without leaving the designer. It is the quickest way to learn what an expression looks like after you built it with clicks.

Classic and new designer

Microsoft has been replacing the classic designer with a new one that has a side panel for parameters and a Copilot pane. Everything in this course applies to both. If a menu is missing, look for it in the other designer through the toggle at the top.

Flow checker

Select Flow checker before you test. It lists errors (the flow cannot be saved or run) and warnings (a likely problem, such as a missing retry policy). Fix errors first; read each warning once and decide.

Testing

Choose Test and select how to run.

  • Manually: you trigger the event yourself (send the email, create the item), and the flow runs when it happens. Good for the first real run.
  • Automatically: you re-run using the data from a previous successful run. This saves you from creating new test data each time. It is only available when the flow has run before.

For a scheduled or instant flow, the test is just a run now.

While a test runs the designer draws a green check or a red cross on each card and shows run time. Cards that were skipped appear greyed out.

Run history

On the flow's details page the 28-day run history lists runs with start time, duration and status: Succeeded, Failed, Running, Cancelled or Skipped.

Open one run to see every step with its inputs and outputs. This is the single most useful debugging tool: you can see the exact value an expression produced.

Run history is kept for 28 days. If you must keep evidence longer, write your own log (for example to a SharePoint list) from inside the flow.

From a run you can:

  • Resubmit a failed trigger run to try again with the same trigger data, after you fixed the cause.
  • Cancel a run that is waiting or running too long.
  • Copy an output and paste it as a sample in Parse JSON (see Arrays, objects and JSON).

Reading a failure

A failed card shows an error code, such as NotFound, Forbidden, BadRequest or InvalidTemplate, and a message. Learn this small table.

Status code Typical meaning
400 BadRequest Input is malformed or missing (check the fields)
401 Unauthorized Connection expired or wrong credentials
403 Forbidden You lack permission on the target
404 NotFound The file, item or path does not exist
429 TooManyRequests Throttled; retry after a pause
5xx Service problem; usually retried automatically

An InvalidTemplate error comes from an expression and names the action and sometimes the position. You will fix those in Expressions first steps.

A stand-in for run history

The next Python stand-in models a run history list and summarizes it. It is not Power Automate, but it shows the questions you ask of real history: how often does it fail, and why?

Python Stand-in: summarize a made-up run history

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

Two failures with NotFound and one Unauthorized tell you where to look first: probably a path that does not exist and a connection that needs re-signing in.

A debugging routine

Follow the same order every time.

  1. Open the failed run and find the first red card.
  2. Read the error message and code.
  3. Compare the card's inputs with what you expected. Most bugs are a wrong or empty input.
  4. Open the previous card's outputs to see where the bad value came from.
  5. Fix, Save, resubmit.
Python Stand-in: walk back to the first failing step

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

Habits that save time

  • Rename cards as you build; the run history shows card names, not types.
  • Use Add a comment for non-obvious expressions.
  • Keep one test email, item or file that you can reuse with Automatically tests.
  • Turn a flow off (not delete) while you edit a live flow with side effects, then test with a copy.

Check your understanding

Quick check ยท 4 questions

of 4 answered You scored / 4 ยท perfect!

  1. 1. Where do you see the exact value an action received and returned?

    Each card in a run shows its inputs and outputs, which is the main debugging aid.
  2. 2. How long is run history kept?

    Run history is kept for 28 days, so log important facts yourself.
  3. 3. What does a 401 status in a failed card usually point to?

    401 means unauthorized, commonly an expired sign-in on the connection.
  4. 4. What does the Automatically test option do?

    It reuses earlier trigger data so you can test without producing a new event.

Summary

  • The designer shows steps as cards; use Rename, Peek code, Settings and the Flow checker.
  • Test manually first, then automatically with earlier trigger data.
  • Run history (28 days) lets you inspect inputs and outputs and resubmit failed runs.
  • Debug from the first red card backwards, reading status codes.

Next: Environments, solutions and the maker portal