Files
sencho/backend/src/__tests__/DatabaseService.autoHeal.test.ts
T
Anso 5bb4b01953 feat: auto-heal policies for unhealthy containers (#671)
* feat(db): add auto_heal_policies and auto_heal_history schema and CRUD

Adds two new SQLite tables (auto_heal_policies, auto_heal_history) to
DatabaseService.initSchema() and exposes CRUD methods: getAutoHealPolicies,
getAutoHealPolicy, addAutoHealPolicy, updateAutoHealPolicy,
deleteAutoHealPolicy, recordAutoHealHistory, getAutoHealHistory,
incrementConsecutiveFailures, resetConsecutiveFailures, setPolicyEnabled.
Also adds AutoHealPolicy and AutoHealHistoryEntry TypeScript interfaces.

* feat(events): track health-status duration and expose state accessors

- Add healthStatus and unhealthySince fields to InternalContainerState
- onHealthStatus now records unhealthySince timestamp on first transition
  to unhealthy, and clears it when the container recovers or restarts
- onStart resets both fields so a restarted container begins from 'starting'
- Add listContainerStates() and getContainerState() public accessors for
  use by the upcoming AutoHealService evaluator

* fix(auto-heal): key allowlist in updateAutoHealPolicy, cascade delete, extract ContainerHealthSnapshot

* feat: add AutoHealService evaluator singleton

Polls every 30 s, matches containers to enabled policies via Compose
labels, and restarts containers that have been unhealthy beyond the
configured threshold. Enforces cooldown, per-hour rate cap, and
recent-user-action suppression; auto-disables policies after repeated
consecutive failures. Also adds DockerEventManager.getService() accessor
required by the evaluator.

* fix(auto-heal): prune stale restartTimestamps, guard undefined policy id

- Prune restartTimestamps entries for containers no longer running after
  each container list fetch, preventing unbounded map growth from dead
  container IDs.
- Guard against policies with undefined id at the start of the per-policy
  loop; warn and skip rather than proceed with a non-null assertion.
- Extract handleAutoDisable private helper to bring executeHeal under 30
  lines and isolate the auto-disable side-effect sequence.
- Move ContainerInfo type to module scope.

* feat: add auto-heal API routes and wire AutoHealService lifecycle

Registers five REST endpoints under /api/auto-heal/policies (list, create,
patch, delete, history) with requirePaid + requireAdmin guards and Zod
validation. Wires AutoHealService.start()/stop() into the server startup
and graceful-shutdown blocks alongside MonitorService.

* test: add AutoHealService and DatabaseService auto-heal unit tests

- 15 unit tests for AutoHealService.shouldHeal covering all decision branches
  (healthy state, duration threshold, user-action suppression, cooldown,
  rate limiting, and correct skipReason values)
- 13 integration tests for DatabaseService auto-heal CRUD: policy round-trip,
  stack-name filter, partial update, cascade delete, history ordering/limit,
  consecutive failure counters, and setPolicyEnabled toggle

* fix: log AutoHealService shutdown errors consistently

* fix(api): requireAdmin-first guard order and try/catch on auto-heal routes

* feat(ui): add StackAutoHealSheet component

* feat(ui): add Auto-Heal context menu item to EditorLayout

* fix(ui): StackAutoHealSheet label, token, a11y, and useEffect fixes

- Rename 'All services in stack' to 'All services' in combobox options and placeholder
- Replace text-green-600 with text-success design token in actionColorClass
- Add htmlFor/id pairs to all four numeric form inputs for accessibility
- Inline fetch logic into useEffect, removing stale closure risk and eslint-disable comment
- Remove now-unused fetchPolicies and fetchServices standalone functions
- Update 'Auto-disable after' label to 'Auto-disable after (failures)' for clarity
- Add toast.error in policy fetch failure path; services fetch silently skips as before

* docs: add auto-heal-policies feature documentation

* test(e2e): add auto-heal policies CRUD spec

* fix(docs): correct auto-heal-policies nav position in docs.json
2026-04-17 22:45:06 -04:00

272 lines
10 KiB
TypeScript

/**
* Integration tests for DatabaseService auto-heal CRUD.
*
* Uses a real temp SQLite database (same pattern as database-metrics.test.ts).
* All operations are tested against live SQL; no mocking.
*/
import { describe, it, expect, beforeAll, afterAll } from 'vitest';
import { setupTestDb, cleanupTestDb } from './helpers/setupTestDb';
let tmpDir: string;
let db: any;
const makePolicy = (overrides: Record<string, unknown> = {}) => {
const now = Date.now();
return {
stack_name: 'teststack',
service_name: null,
unhealthy_duration_mins: 5,
cooldown_mins: 10,
max_restarts_per_hour: 3,
auto_disable_after_failures: 5,
enabled: 1,
consecutive_failures: 0,
last_fired_at: 0,
created_at: now,
updated_at: now,
...overrides,
};
};
beforeAll(async () => {
tmpDir = await setupTestDb();
const { DatabaseService } = await import('../services/DatabaseService');
db = DatabaseService.getInstance();
});
afterAll(() => {
cleanupTestDb(tmpDir);
});
describe('DatabaseService - auto-heal policy CRUD', () => {
it('addAutoHealPolicy + getAutoHealPolicy round-trip preserves all fields', () => {
const input = makePolicy({
stack_name: 'roundtrip-stack',
service_name: 'web',
unhealthy_duration_mins: 3,
cooldown_mins: 15,
max_restarts_per_hour: 5,
auto_disable_after_failures: 10,
enabled: 1,
consecutive_failures: 0,
last_fired_at: 0,
});
const created = db.addAutoHealPolicy(input);
expect(created.id).toBeDefined();
expect(typeof created.id).toBe('number');
const fetched = db.getAutoHealPolicy(created.id);
expect(fetched).toBeDefined();
expect(fetched.stack_name).toBe('roundtrip-stack');
expect(fetched.service_name).toBe('web');
expect(fetched.unhealthy_duration_mins).toBe(3);
expect(fetched.cooldown_mins).toBe(15);
expect(fetched.max_restarts_per_hour).toBe(5);
expect(fetched.auto_disable_after_failures).toBe(10);
expect(fetched.enabled).toBe(1);
expect(fetched.consecutive_failures).toBe(0);
expect(fetched.last_fired_at).toBe(0);
expect(fetched.created_at).toBe(input.created_at);
});
it('getAutoHealPolicies without filter returns all policies', () => {
const before = db.getAutoHealPolicies().length;
db.addAutoHealPolicy(makePolicy({ stack_name: 'all-stack-a' }));
db.addAutoHealPolicy(makePolicy({ stack_name: 'all-stack-b' }));
const after = db.getAutoHealPolicies();
expect(after.length).toBe(before + 2);
});
it('getAutoHealPolicies with stackName filter returns only matching rows', () => {
db.addAutoHealPolicy(makePolicy({ stack_name: 'filter-stack-x' }));
db.addAutoHealPolicy(makePolicy({ stack_name: 'filter-stack-y' }));
const xPolicies = db.getAutoHealPolicies('filter-stack-x');
const yPolicies = db.getAutoHealPolicies('filter-stack-y');
expect(xPolicies.length).toBeGreaterThanOrEqual(1);
expect(xPolicies.every((p: any) => p.stack_name === 'filter-stack-x')).toBe(true);
expect(yPolicies.length).toBeGreaterThanOrEqual(1);
expect(yPolicies.every((p: any) => p.stack_name === 'filter-stack-y')).toBe(true);
});
it('updateAutoHealPolicy partial update changes only specified fields', () => {
const created = db.addAutoHealPolicy(makePolicy({ stack_name: 'update-stack', cooldown_mins: 10 }));
db.updateAutoHealPolicy(created.id, { cooldown_mins: 30, max_restarts_per_hour: 7 });
const updated = db.getAutoHealPolicy(created.id);
// Changed fields
expect(updated.cooldown_mins).toBe(30);
expect(updated.max_restarts_per_hour).toBe(7);
// Unchanged fields must be preserved
expect(updated.stack_name).toBe('update-stack');
expect(updated.unhealthy_duration_mins).toBe(5);
// updated_at should be refreshed
expect(updated.updated_at).toBeGreaterThanOrEqual(created.updated_at);
});
it('deleteAutoHealPolicy removes the policy row', () => {
const created = db.addAutoHealPolicy(makePolicy({ stack_name: 'delete-stack' }));
expect(db.getAutoHealPolicy(created.id)).toBeDefined();
db.deleteAutoHealPolicy(created.id);
expect(db.getAutoHealPolicy(created.id)).toBeUndefined();
});
it('deleteAutoHealPolicy cascades and removes history rows', () => {
const created = db.addAutoHealPolicy(makePolicy({ stack_name: 'cascade-stack' }));
const policyId: number = created.id;
// Write a history entry tied to this policy
db.recordAutoHealHistory({
policy_id: policyId,
stack_name: 'cascade-stack',
service_name: null,
container_name: 'cascade-stack-web-1',
container_id: 'cascade-abc',
action: 'restarted',
reason: 'test cascade',
success: 1,
error: null,
timestamp: Date.now(),
});
const beforeDelete = db.getAutoHealHistory(policyId);
expect(beforeDelete.length).toBe(1);
db.deleteAutoHealPolicy(policyId);
const afterDelete = db.getAutoHealHistory(policyId);
expect(afterDelete.length).toBe(0);
});
});
describe('DatabaseService - auto-heal history', () => {
it('recordAutoHealHistory + getAutoHealHistory orders results by timestamp DESC', () => {
const policy = db.addAutoHealPolicy(makePolicy({ stack_name: 'history-order-stack' }));
const policyId: number = policy.id;
const baseTs = Date.now();
db.recordAutoHealHistory({
policy_id: policyId,
stack_name: 'history-order-stack',
service_name: null,
container_name: 'history-order-stack-web-1',
container_id: 'hist-001',
action: 'restarted',
reason: 'first',
success: 1,
error: null,
timestamp: baseTs,
});
db.recordAutoHealHistory({
policy_id: policyId,
stack_name: 'history-order-stack',
service_name: null,
container_name: 'history-order-stack-web-1',
container_id: 'hist-001',
action: 'skipped_cooldown',
reason: 'second',
success: 0,
error: null,
timestamp: baseTs + 1_000,
});
const history = db.getAutoHealHistory(policyId);
expect(history.length).toBe(2);
// Most recent first
expect(history[0].reason).toBe('second');
expect(history[1].reason).toBe('first');
});
it('getAutoHealHistory respects the limit parameter', () => {
const policy = db.addAutoHealPolicy(makePolicy({ stack_name: 'history-limit-stack' }));
const policyId: number = policy.id;
const baseTs = Date.now();
for (let i = 0; i < 10; i++) {
db.recordAutoHealHistory({
policy_id: policyId,
stack_name: 'history-limit-stack',
service_name: null,
container_name: 'history-limit-stack-web-1',
container_id: 'hist-limit',
action: 'restarted',
reason: `entry-${i}`,
success: 1,
error: null,
timestamp: baseTs + i,
});
}
const limited = db.getAutoHealHistory(policyId, 3);
expect(limited.length).toBe(3);
// Should contain the 3 most recent entries (highest timestamps)
expect(limited[0].reason).toBe('entry-9');
expect(limited[1].reason).toBe('entry-8');
expect(limited[2].reason).toBe('entry-7');
});
});
describe('DatabaseService - consecutive failure counters', () => {
it('incrementConsecutiveFailures increments the counter each call', () => {
const policy = db.addAutoHealPolicy(makePolicy({ stack_name: 'fail-inc-stack', consecutive_failures: 0 }));
const id: number = policy.id;
db.incrementConsecutiveFailures(id);
expect(db.getAutoHealPolicy(id).consecutive_failures).toBe(1);
db.incrementConsecutiveFailures(id);
expect(db.getAutoHealPolicy(id).consecutive_failures).toBe(2);
});
it('resetConsecutiveFailures sets the counter back to 0', () => {
const policy = db.addAutoHealPolicy(makePolicy({ stack_name: 'fail-reset-stack', consecutive_failures: 0 }));
const id: number = policy.id;
db.incrementConsecutiveFailures(id);
db.incrementConsecutiveFailures(id);
expect(db.getAutoHealPolicy(id).consecutive_failures).toBe(2);
db.resetConsecutiveFailures(id);
expect(db.getAutoHealPolicy(id).consecutive_failures).toBe(0);
});
});
describe('DatabaseService - setPolicyEnabled', () => {
it('setPolicyEnabled(id, false) sets enabled to 0', () => {
const policy = db.addAutoHealPolicy(makePolicy({ stack_name: 'enable-toggle-stack', enabled: 1 }));
const id: number = policy.id;
db.setPolicyEnabled(id, false);
expect(db.getAutoHealPolicy(id).enabled).toBe(0);
});
it('setPolicyEnabled(id, true) sets enabled to 1', () => {
const policy = db.addAutoHealPolicy(makePolicy({ stack_name: 'enable-on-stack', enabled: 0 }));
const id: number = policy.id;
db.setPolicyEnabled(id, true);
expect(db.getAutoHealPolicy(id).enabled).toBe(1);
});
it('getAutoHealPolicies returns only enabled policies after filtering', () => {
const enabledPolicy = db.addAutoHealPolicy(makePolicy({ stack_name: 'enabled-filter-stack', enabled: 1 }));
const disabledPolicy = db.addAutoHealPolicy(makePolicy({ stack_name: 'disabled-filter-stack', enabled: 0 }));
const allPolicies = db.getAutoHealPolicies();
const enabledIds = allPolicies.filter((p: any) => p.enabled === 1).map((p: any) => p.id);
const disabledIds = allPolicies.filter((p: any) => p.enabled === 0).map((p: any) => p.id);
expect(enabledIds).toContain(enabledPolicy.id);
expect(disabledIds).toContain(disabledPolicy.id);
expect(enabledIds).not.toContain(disabledPolicy.id);
});
});