Files
0pt/app/api/health/route.ts
T
admin a88ad04867
Security Scan (SIAX Cloud) / security (push) Successful in 21s
CI (SIAX Cloud) / security (push) Successful in 21s
CI (SIAX Cloud) / quality (push) Successful in 41s
CI (SIAX Cloud) / security (pull_request) Successful in 22s
Security Scan (SIAX Cloud) / security (pull_request) Successful in 21s
CI (SIAX Cloud) / quality (pull_request) Successful in 41s
fix(health): report real per-dependency status instead of hardcoded ok:true
BACKLOG P0-3: /api/health only ever checked *presence* of DATABASE_URL/
REDIS_URL and always returned ok:true regardless of whether anything was
reachable -- ARCHITECTURE.md's own words, "an uptime monitor watching HTTP
200 would show green for a dying instance".

lib/health-checks.ts adds real dependency probes:
  - db: a live `select 1` round trip (not just DATABASE_URL presence)
  - engine: a real HTTP call to the FastAPI engine's own GET /health
  - stripe: Stripe's documented lightweight balance.retrieve() liveness call
  - queue: honestly reports "not_integrated" when REDIS_URL is set but
    unused -- no Redis client is wired anywhere in this repo, so reporting
    "ok" would fabricate a check that doesn't exist

/api/health stays a liveness check (always 200 while the process can
respond; the body's `ok` field is the honest signal). New GET /api/ready
fails closed with 503 when the load-bearing dependency (db) isn't both
configured and reachable -- the "no /ready route" gap.

contracts/openapi.yaml, docs/architecture/ARCHITECTURE.md, and docs/slo.md
updated to match. 17 new unit tests (dependency-injected fakes, no real
network/DB) + 5 route-level tests.
2026-08-11 13:49:23 +02:00

44 lines
1.8 KiB
TypeScript

import { NextResponse } from "next/server";
import { checkAllDependencies, hasDegradedDependency } from "@/lib/health-checks";
// GET /api/health — liveness (docs/04, BACKLOG P0-3).
//
// Fixed finding: this used to hardcode `ok: true` and only ever check
// *presence* of DATABASE_URL/REDIS_URL, never whether anything was actually
// reachable -- "an uptime monitor watching HTTP 200 would show green for a
// dying instance" (docs/architecture/ARCHITECTURE.md). Each dependency below
// now reports a real state: "ok" (live round trip succeeded), "error"
// (configured but unreachable -- see `detail`), "not_configured", or
// "not_integrated" (env var present but no client code reads it -- honest
// about `queue`, which has no wired Redis client in this repo).
//
// This route stays a LIVENESS check: it always answers 200 as long as the
// process itself can run the handler (that's what "liveness" means -- an
// external dependency outage is not a reason to restart this instance). The
// body's `ok` field is the honest signal instead: false whenever a
// dependency that IS configured is failing. For a readiness gate that DOES
// fail closed on missing/broken dependencies, see GET /api/ready.
export async function GET() {
const deps = await checkAllDependencies();
const degraded = hasDegradedDependency(deps);
return NextResponse.json({
ok: !degraded,
service: "0pt-web",
db: deps.db.status,
queue: deps.queue.status,
engine: deps.engine.status,
stripe: deps.stripe.status,
...(degraded
? {
detail: Object.fromEntries(
Object.entries(deps)
.filter(([, v]) => v.status === "error")
.map(([k, v]) => [k, v.detail]),
),
}
: {}),
ts: new Date().toISOString(),
});
}