Environments

The nijam_environment ini option tags each run with the target it ran against - any string you like: staging, production, pr-preview, and so on. The dashboard then offers an environment filter so you can scope runs (and flakiness) to one target.

Setting it

Set it in pytest.ini, or wire it to the NIJAM_ENVIRONMENT env var so each pipeline reports its own target (the env var wins when both are set):

pytest.ini

[pytest]
nijam_project_id = <your-project-uuid>
nijam_environment = staging

# or override per pipeline
export NIJAM_ENVIRONMENT=production

It’s free-form - Nijam doesn’t validate it against a fixed set. Pick a convention and stick to it so the filter stays tidy.

Unset runs

Runs reported without an environment are grouped under Unset in the dashboard. That’s a feature, not an error - you can always tell which runs carried no environment info, and filter them out.

Why it matters for flakiness

A test that’s “flaky everywhere” and one that’s “flaky only on staging” are different problems. Tagging environments lets Nijam - and you - separate the two when reading a test’s history. See Flakiness.