Reading Results
Results appear in the panel on the right of the Tests tab.
- Latest run shows the run you just started. Each suite appears as soon as it finishes, so you don't wait for the slowest one.
- Run history lists earlier runs. Click one to see it again. Clear deletes the history (the suites stay).
What each result means
| Result | Meaning |
|---|---|
| ✓ passed | The route ran and every check held |
| ✗ failed | The route ran, but at least one check did not hold |
| ✗ error | The suite could not finish: setup failed, the route crashed, or the request could not be sent |
| ⚠ timeout | Setup, a workflow's input script or loader, or the route or workflow took longer than its time limit and was stopped |
A run passes when all its suites pass.
Why did it fail?
Click a suite's row to open it. You'll see:
- Every check, with a ✓ or ✗. A failed check says what it expected and what it got, for example
expected 500 to be 201. For simple checks, a path that isn't in the response says so and lists what is (details). Checks from hooks come first, labelled with their block. - The error, if the suite hit one (for example
Setup failed: duplicate email). - The response the route sent: status, headers and body, so you can compare them with what you expected.
A suite stopped for taking too long has no checks to show, because it never got that far.
See the run block by block
Every test run is recorded, even when the route's Recording switch is off. Next to a case, View trace opens the run it made on the canvas: the path it took, and each block's input, output and error. It is the fastest way to find the block that failed.
Blocks a hook skipped or changed carry a Mocked badge. Once a trace is older than the recording retention (30 days by default), the results show Trace expired instead. See Execution Recording.
Teardown warnings
If a suite's teardown fails, the suite keeps its result and shows ⚠ Teardown failed next to its name. Open it to see the reason. When the run finishes, a notice also tells you how many suites had this problem, because their test data may still be in the database.
Common problems
| You see | What it means | What to do |
|---|---|---|
| "No development worker is running" | Suites run on the development worker, and none is running | Start a worker with FLUXIFY_ENV=development, then run again |
| "No result from a worker within …" | The worker didn't answer in time, for example because it restarted mid-run | Run again. If it keeps happening, raise the suite's time limits |
| "Custom JS made no t.expect(...) checks" | A Custom JS check doesn't call t.expect | Rewrite it with t.expect(...). See Custom JS checks |
| "Setup failed: …" | The setup block threw an error | Open the setup block and fix it. The route was skipped |
| "setup timed out" / "suite timed out" / "case 3 timed out" / "input timed out" | A phase took longer than its limit. Setup and the input loader use the limits on the suite; the route or workflow uses its own time limit | Raise the limit, or find what is slow (a hook, a slow service) |
| The request doesn't reach the route | A path value is missing, so the address doesn't match the route | Fill every Path parameter in the Request tab |
| A suite passes alone but fails with Run all | Suites running at the same time see each other's data | Make test data unique with testsuite.runId, or tick Run alone. See Run alone |
