tidaldb/tests/e2e/features/03-auth-boundary.spec.ts
jordan 15f6b11187 test(e2e): Playwright evidence harness for the deploy-verification runbook
Turns docs/runbooks/deploy-verification.md from prose into 32 executable checks
against the live orchard9-k3sf cluster, and it found real defects on its first
run — including in the runbook it verifies.

WHY PLAYWRIGHT, HONESTLY
tidalDB serves zero HTML (no text/html, no Html(), 10 JSON routes), so this uses
Playwright in three distinct roles rather than pretending there is a UI:
  * request fixture as a real HTTP client for DNS/TLS/auth/quorum/404;
  * a browser for the only genuine screens in the chain, Grafana;
  * a test harness for cluster-plane checks with no HTTP surface, shelling out
    to kubectl and attaching the real transcript as evidence.

WHAT IT CAUGHT
  * The runbook asserted the operator/data credential split was "not active yet
    - requires an image roll". globalSetup read the live image and the live
    secret; a probe returned data->403, admin->200. It had been enforcing the
    whole time. Section 9 rewritten. (BUG-001)
  * docs/ops/grafana-tidaldb.json shipped datasource uid ${DS_PROMETHEUS} - a
    Grafana export-for-sharing placeholder with no __inputs block to resolve it.
    Under ConfigMap provisioning every panel queried a datasource that did not
    exist, so the whole board was blank. The API said "loaded" and I had only
    ever checked the API. 41 refs fixed here, 58 across the fleet ConfigMap,
    which was also blanking the postgres and redis dashboards. (BUG-007)
  * Stat panels used calcs "lastNonNull". Grafana's reducer is "lastNotNull", so
    no value was ever computed and Cluster health / Reseed pending / Indexed
    vectors rendered as empty boxes. I chased panel width and then panel height
    before comparing against a working stat panel elsewhere in the same Grafana.
    A spelling error wearing a layout bug's clothes. (BUG-009)
  * The namespace variable defaulted to All, so cluster panels silently included
    tidaldb-586b544c8-vpkmw from the superseded standalone deployment. Latency
    legends read "p50 p50 p50" with no way to tell the nodes apart. Both fixed.
  * "5xx ratio" rendered "No data" as large green text - at a glance a healthy
    value. And Fleet state gave three fields one shared green threshold, so
    reseed_required=1 would have shown GREEN during the exact incident the panel
    exists to surface. Split into three panels with per-field mappings.
  * tidalctl cluster-status exits 2 on a FULLY CONVERGED cluster, because the
    aggregated endpoint reports healthy peers as region=null applied=0
    reachable=false. The runbook claimed `cluster-status && deploy` was a safe
    gate; that claim came from an exit code masked by a shell pipeline. The gate
    can never pass here. Documented, test pins it, engine defect recorded.
    (BUG-005)
  * The deployed image writes ANSI colour into container logs, which the
    collector stores verbatim. Already fixed in logging.rs, not yet rolled;
    pinned as a tripwire. (BUG-006)
  * The runbook's own backup command sorted ALL backups by timestamp and
    selected a restore-canary run: 20 items, one volume, a meaningless pass.
    Now filters on the schedule label the freshness alert actually watches.

DEFECTS FOUND BY LOOKING AT THE SCREENS
Six of the first eight captures were slop and were fixed, not promoted:
230-350px of dead space; a verdict that rendered "exit code 2" in green; the
1600x1800 dashboard scaled into 16:9 until illegible (now clipped to the
evidence band using real element bounds); the dream beat whose caption described
a contradiction the image did not show (now a purpose-built capture holding the
committed doc text, the running image, and the live 403/200 side by side); and a
one-frame blink to bare background at every scene boundary, because Remotion
Sequences do not overlap and both scenes sat at opacity 0 on the boundary frame.

TRIPWIRES IN THE HONEST DIRECTION
Three tests assert what is ABSENT - zero tidaldb_http_* families, JSON_LOGS
unset, plain-text logs - and each carries the message "good news, roll the
runbook section from pending to live". The metric-absence test also asserts the
baseline family count, so "absent" cannot pass for "the scrape failed". That is
the drift that made section 9 stale in the first place.

Regression config uses workers:1 and retries:0 deliberately: a live-cluster
check that only passes on the second attempt has told you something true.

Verified: 32 passed (46.8s); 9 demo captures each asserting before photographing;
tsc clean; render 82.05s 1920x1080 h264, 0 empty frames across 10 boundaries;
every promoted image inspected individually and judged perfect; walk-the-render
ledger complete with no fails.
2026-08-23 14:03:29 -06:00

157 lines
5.7 KiB
TypeScript

/**
* Runbook section 3 — the authentication boundary and a real committed write.
*
* CAP-005 the data plane refuses unauthenticated and wrong credentials
* CAP-006 a quorum-acked write commits end to end
* CAP-007 operator and metrics surfaces are unroutable from the internet
*
* The negative cases are the point. A suite that only proves the happy path has
* proved that the endpoint works, not that it is protected — and this endpoint
* is on the open internet.
*/
import { expect, test } from '@playwright/test';
import { recordJson } from '../support/evidence';
import { PUBLIC_BASE_URL, apiKey } from '../support/env';
/**
* Entity ids for verification writes live in a reserved band so a probe is
* never mistaken for corpus data by someone reading the store later.
*/
const VERIFICATION_ENTITY_BASE = 999_000_000;
/**
* Paths published through the ingress allowlist in k8s/cluster/ingress.yaml.
* Everything else must be rejected by the gateway before it reaches the app.
*/
const UNPUBLISHED_PATHS = [
'/cluster/status',
'/cluster/members',
'/metrics',
'/openapi.json',
] as const;
test.describe('section 3 — authentication boundary', () => {
test('the same path is refused without a credential, refused with a wrong one, and served with the right one', async ({
playwright,
}, testInfo) => {
const anonymous = await playwright.request.newContext();
const wrong = await playwright.request.newContext({
extraHTTPHeaders: { authorization: 'Bearer definitely-not-the-key' },
});
const authorized = await playwright.request.newContext({
extraHTTPHeaders: { authorization: `Bearer ${apiKey()}` },
});
try {
const target = `${PUBLIC_BASE_URL}/search?query=verification&limit=1`;
const [anonymousResponse, wrongResponse, authorizedResponse] = [
await anonymous.get(target),
await wrong.get(target),
await authorized.get(target),
];
const observedStatuses = {
noCredential: anonymousResponse.status(),
wrongCredential: wrongResponse.status(),
correctCredential: authorizedResponse.status(),
};
await recordJson(testInfo, 'auth-matrix', observedStatuses);
// 401 for both denials. A 200 on the wrong bearer would be the highest
// severity finding this suite can produce; a 500 would mean the key was
// parsed and then mishandled.
expect(observedStatuses.noCredential, 'anonymous read must be refused').toBe(401);
expect(observedStatuses.wrongCredential, 'wrong bearer must be refused').toBe(401);
expect(observedStatuses.correctCredential, 'correct bearer must be served').toBe(200);
// A 200 with an empty body would mean the credential was accepted but the
// engine returned nothing — proof of auth, not proof of service.
const payload = await authorizedResponse.json();
await recordJson(testInfo, 'authorized-search-shape', {
keys: Object.keys(payload as Record<string, unknown>),
});
expect(
Object.keys(payload as Record<string, unknown>).length,
'an authorized read must return a real payload',
).toBeGreaterThan(0);
} finally {
await anonymous.dispose();
await wrong.dispose();
await authorized.dispose();
}
});
test('a quorum-acked write is committed, not merely accepted', async ({
playwright,
}, testInfo) => {
const context = await playwright.request.newContext({
baseURL: PUBLIC_BASE_URL,
extraHTTPHeaders: { authorization: `Bearer ${apiKey()}` },
});
try {
const entityId = VERIFICATION_ENTITY_BASE + 99;
const response = await context.post('/items', {
headers: { 'content-type': 'application/json', 'x-tidal-ack': 'quorum' },
data: {
entity_id: entityId,
metadata: {
title: 'deploy verification probe',
category: 'verification',
},
},
});
const body = await response.text();
await recordJson(testInfo, 'quorum-write', {
entityId,
status: response.status(),
body: body.slice(0, 300),
});
// 201 specifically. 202 would mean the write was accepted for later
// replication — that is exactly the weaker guarantee this check exists to
// rule out, so it must not be tolerated as "close enough".
expect(
response.status(),
'a quorum-ack write must return 201 (committed by quorum), not 202 (accepted)',
).toBe(201);
} finally {
await context.dispose();
}
});
test('operator and metrics surfaces are rejected by the gateway, not merely unauthorized', async ({
playwright,
}, testInfo) => {
// Deliberately carrying a VALID credential. If a path answered 401 that
// would be tolerable; answering 200 to a valid key on /cluster/status would
// mean the allowlist has drifted and the operator surface is public.
const context = await playwright.request.newContext({
baseURL: PUBLIC_BASE_URL,
extraHTTPHeaders: { authorization: `Bearer ${apiKey()}` },
});
try {
const statuses: Record<string, number> = {};
for (const path of UNPUBLISHED_PATHS) {
statuses[path] = (await context.get(path)).status();
}
await recordJson(testInfo, 'unpublished-path-statuses', statuses);
for (const path of UNPUBLISHED_PATHS) {
// 404 means Traefik never routed it. 401 would mean the request reached
// the application and only the credential check stopped it — the
// gateway allowlist would have drifted.
expect(
statuses[path],
`${path} must be unroutable (404) from the internet, not merely unauthorized`,
).toBe(404);
}
} finally {
await context.dispose();
}
});
});