Files
sencho/backend/src/__tests__/fleet-prune-df-timeout.test.ts
T
Anso a51547a158 fix(monitor): decouple janitor disk-usage check from 30s cycle (F-6) (#1164)
* fix(monitor): decouple janitor disk-usage check from 30s cycle (F-6)

`docker system df` (called by the MonitorService janitor check) can take
30+ seconds on Docker Desktop with many volumes. Running it on the 30s
evaluate cycle compounded with the per-container stats fan-out and pushed
the cycle to 140s+, blocking subsequent monitoring work.

This change:

- Moves the janitor disk-usage check into its own 15-minute cycle with
  a tight 8s timeout. A circuit breaker opens after 3 consecutive
  timeouts (60-minute cooldown) so a sick daemon stops pinning Dockerode
  sockets every tick. The first janitor tick is deferred 45 seconds past
  boot to avoid head-of-line collision with the initial monitor cycle's
  stats fan-out.
- Adds a paired 8s `withTimeout` wrap to the admin prune-estimate routes
  (`/api/system/prune/estimate` and the dry-run path of
  `/api/system/prune/system`) so a slow df does not hang the admin tab.
  Both routes respond 503 with code `docker_df_slow` on timeout.
- Factors `withTimeout` and `TimeoutError` into `utils/withTimeout.ts`
  so the route layer does not have to import from a service module.
- Adds 10 unit tests covering the decoupling guardrail, breaker
  open/close, cooldown, threshold gate, the 100 MB reclaimable floor,
  re-entrancy, recovery logging, non-timeout error handling, and the
  full timer-cleanup contract of `stop()`.
- Adds 4 integration tests for the prune routes covering the 503
  timeout response, the success path, and the non-timeout 5xx path.

* fix(fleet,monitor): extend F-6 timeout to fleet prune routes; close breaker-recovery log gap

Codex audit findings on PR #1164:

Major. The fleet routes that fan out prune-estimate work on local nodes
(`POST /api/fleet/labels/fleet-prune` dry-run path and
`POST /api/fleet/prune/estimate`) called `estimateSystemReclaim` without
a timeout, so a slow local Docker daemon could still hang the fleet
admin tab even though the system-maintenance routes were already
bounded. Wrap both call sites with the shared `withTimeout(..., 8s)`
and surface a "Docker daemon is busy" message via the per-target and
per-node error channels the routes already used for other failures.
The destructive (non-dry-run) prune path stays unwrapped because it
calls `pruneSystem` / `pruneManagedOnly`, not `df`.

Minor. The janitor circuit breaker zeroed `janitorConsecutiveTimeouts`
when it opened, so a successful call after a full breaker-open cooldown
slipped past the `if (counter > 0)` recovery-log branch and never
emitted `[Monitor] Janitor disk-usage check recovered`. The operator
observability signal was missing exactly when it mattered most.
Extend the predicate to also trip on `janitorBreakerUntil > 0` (which
stays set to its past timestamp after cooldown until the next success
clears it), so recovery logs symmetrically for both partial-failure
and post-breaker recovery paths. Added a dedicated test.

Three new integration tests cover the fleet routes (timeout, success,
and the estimate endpoint's per-node unreachable shape).
2026-05-23 00:04:05 -04:00

101 lines
4.1 KiB
TypeScript

/**
* F-6 regression: fleet routes that call estimateSystemReclaim on local
* nodes must also bound the slow `docker system df` call (8s) and surface
* a recognizable timeout message to the operator, matching the
* /api/system/prune/estimate behavior.
*
* Covers:
* - POST /api/fleet/labels/fleet-prune with dryRun: true
* - POST /api/fleet/prune/estimate
*
* Uses real timers because supertest dispatches lazily and the in-route
* `withTimeout` setTimeout cannot be advanced via vi.useFakeTimers from
* outside the request lifecycle.
*/
import { describe, it, expect, beforeAll, afterAll, afterEach, vi } from 'vitest';
import request from 'supertest';
import jwt from 'jsonwebtoken';
import { setupTestDb, cleanupTestDb, TEST_USERNAME, TEST_JWT_SECRET } from './helpers/setupTestDb';
let tmpDir: string;
let app: import('express').Express;
let authHeader: string;
let DockerController: typeof import('../services/DockerController').default;
let FileSystemService: typeof import('../services/FileSystemService').FileSystemService;
let activeBulkActions: typeof import('../routes/labels').activeBulkActions;
beforeAll(async () => {
tmpDir = await setupTestDb();
({ app } = await import('../index'));
({ default: DockerController } = await import('../services/DockerController'));
({ FileSystemService } = await import('../services/FileSystemService'));
({ activeBulkActions } = await import('../routes/labels'));
// 10-minute expiry survives the file even with two ~8s timeout tests.
const token = jwt.sign({ username: TEST_USERNAME }, TEST_JWT_SECRET, { expiresIn: '10m' });
authHeader = `Bearer ${token}`;
});
afterAll(() => cleanupTestDb(tmpDir));
afterEach(() => {
vi.restoreAllMocks();
activeBulkActions.clear();
});
function stubLocalEstimate(impl: () => Promise<{ reclaimableBytes: number }>) {
vi.spyOn(DockerController, 'getInstance').mockReturnValue({
estimateSystemReclaim: vi.fn().mockImplementation(impl),
estimateManagedReclaim: vi.fn().mockResolvedValue({ reclaimableBytes: 0 }),
} as unknown as ReturnType<typeof DockerController.getInstance>);
vi.spyOn(FileSystemService.prototype, 'getStacks').mockResolvedValue([]);
}
describe('Fleet prune routes bound docker df at 8s on local nodes (F-6)', () => {
it('POST /api/fleet/labels/fleet-prune dry-run surfaces a busy-daemon error on local timeout', async () => {
stubLocalEstimate(() => new Promise(() => { /* never resolves */ }));
const t0 = Date.now();
const res = await request(app)
.post('/api/fleet/labels/fleet-prune')
.set('Authorization', authHeader)
.send({ targets: ['volumes'], scope: 'all', dryRun: true });
const elapsed = Date.now() - t0;
expect(res.status).toBe(200);
const local = res.body.results[0];
expect(local.reachable).toBe(true);
expect(local.targets[0].success).toBe(false);
expect(local.targets[0].error).toMatch(/Docker daemon is busy/);
expect(elapsed).toBeGreaterThanOrEqual(7_500);
expect(elapsed).toBeLessThan(15_000);
}, 20_000);
it('POST /api/fleet/prune/estimate marks the local node unreachable with a busy-daemon error on timeout', async () => {
stubLocalEstimate(() => new Promise(() => { /* never resolves */ }));
const res = await request(app)
.post('/api/fleet/prune/estimate')
.set('Authorization', authHeader)
.send({ targets: ['volumes'], scope: 'all' });
expect(res.status).toBe(200);
expect(Array.isArray(res.body.perNode)).toBe(true);
const local = res.body.perNode[0];
expect(local.reachable).toBe(false);
expect(local.error).toMatch(/Docker daemon is busy/);
}, 20_000);
it('fleet-prune dry-run succeeds normally when estimateSystemReclaim resolves quickly', async () => {
stubLocalEstimate(() => Promise.resolve({ reclaimableBytes: 256 }));
const res = await request(app)
.post('/api/fleet/labels/fleet-prune')
.set('Authorization', authHeader)
.send({ targets: ['volumes'], scope: 'all', dryRun: true });
expect(res.status).toBe(200);
const local = res.body.results[0];
expect(local.targets[0]).toMatchObject({ target: 'volumes', success: true, reclaimedBytes: 256, dryRun: true });
});
});