Executions¶

An execution is one run of one test case, recorded step by step. Executions are allowed only against an approved Test Protocol and require the Tester role.

Executing a manual test¶

1. Open the test case and choose Execute.
2. For each step, record the actual result and Pass or Fail.
3. Attach evidence to a step where needed (screenshots, exports, printouts).
4. If the test must pass in named environments, select the environment you ran it in.
5. Sign the execution. See Signatures.

A signed execution is write-once: it cannot be edited or deleted.

Evidence¶

Each evidence file is stored with its SHA-256 hash and served only to members of the system. Capture screen on a step lets you pick the desktop, a window or a browser tab and attaches a PNG as evidence; this needs a secure (HTTPS) connection.

Independence¶

For Tier 1 requirements, a manual test must be executed by someone other than its author. Otherwise the RTM shows executor-not-independent.

Failed steps¶

A failed step offers Raise deviation, pre-filled from the step. Every failed execution needs a closed deviation before the VSR can be signed. See Deviations.

Recording pytest-gxp results by import¶

If you run the qualification yourself, import the pytest-gxp report with Record results. This signs one execution per test and links evidence by hash from evidence_manifest.json. Alternatively let the app run the suite: see Automated runs.