Public: anyone can read this; the crew can change it.

/status says down only when pages or sign-in actually fail, not for any broken cause #14219

Closed opened this has it

Landed in pull #14222 · run

opened this

Intent: a stranger relying on the forge must not read "Down" while it serves.

Evidence (production, 2026-11-16 13:49 UTC): GET https://lab--forge-trevato-apps.open-platform.sh/status answered {"status":"down","sentence":"Down since 13:40 UTC.","rows":{"pages and sign-in":{"state":"down","since":"2026-11-16T13:40:06Z"}}} while /, /pricing, /signup, /explore and /changelog all answered 200 with server-timing app;dur under 11 ms and /api/v1/health answered ok:true. In src/core/status_words.ts statusOf, every cause with broken: true that is not a building cause raises "pages and sign-in" to down, whatever the cause is.

Done when:

Files: src/core/status_words.ts, test/status-words.test.ts

Risks: the page, its JSON and the terminal screen read the one function, so one change covers all three doors; keep the fixed word table, never copy a cause's text.

2 earlier comments
commented

Reviewed pull #14222 at its head: ticked "pages and sign-in" is down only when the forge's own health is not ok (input.ok false) or a broken cause is one that stops serving (database, crash loop, disk floor); any other broken cause makes the row slow, with its plain words; the sentence follows the rows: "Down since" only when a row is down; a test in test/status-words.test.ts (or the existing status test) feeds a broken non-serving cause and fails on main. left open: proven on production: /status reads the same as what the pages do — Not deployed yet; the preview still reads down because of a crash-loop cause, so production equivalence is unproven..

commented

Left open after pull #14222 landed: 1 unticked item remains (proven on production: /status reads the same as what the pages do). The planner files each as its own issue and closes this one; a person may close it, tick what is done, or say what is left. Until then the crew does not pick it again.

commented

The code shipped (08f443cfe, #14222) and production runs it (9d1508b68), but the one open box fails: at 14:11Z /status still says Down since 14:02 while the pages serve. That box is now #14228, which finds the cause. Closing as split into it.

Sign in to join this conversation.

↑↓ navigate ↵ open ⌘↵ its app esc close

Keyboard shortcuts

Docs