mirror of
https://github.com/Studio-Saelix/sencho.git
synced 2026-09-03 06:07:58 +00:00
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
This commit is contained in:
Anso
@@ -0,0 +1,168 @@
|
||||
/**
|
||||
* Unit tests for AutoHealService.shouldHeal decision logic.
|
||||
*
|
||||
* shouldHeal is private; accessed via type cast (service as any) to avoid
|
||||
* exposing it in production API surface. All tests are pure (no I/O, no
|
||||
* timers) - they exercise the decision function directly.
|
||||
*/
|
||||
import { describe, it, expect, beforeEach } from 'vitest';
|
||||
import { INTENTIONAL_KILL_WINDOW_MS } from '../services/ContainerLifecycleClassifier';
|
||||
import { AutoHealService } from '../services/AutoHealService';
|
||||
|
||||
describe('AutoHealService.shouldHeal', () => {
|
||||
let service: any;
|
||||
|
||||
beforeEach(() => {
|
||||
// Reset singleton so each test starts with a clean restartTimestamps map
|
||||
(AutoHealService as any).instance = undefined;
|
||||
service = AutoHealService.getInstance();
|
||||
});
|
||||
|
||||
const basePolicy = {
|
||||
id: 1,
|
||||
stack_name: 'mystack',
|
||||
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: Date.now(),
|
||||
updated_at: Date.now(),
|
||||
};
|
||||
|
||||
const baseState = {
|
||||
id: 'container123',
|
||||
name: 'mystack-web-1',
|
||||
stackName: 'mystack',
|
||||
healthStatus: 'unhealthy' as const,
|
||||
unhealthySince: Date.now() - 6 * 60_000, // 6 minutes ago (past 5 min threshold)
|
||||
lastKillAt: undefined,
|
||||
};
|
||||
|
||||
it('returns heal:true when all conditions are met', () => {
|
||||
const result = service.shouldHeal(baseState, basePolicy, 'container123', Date.now());
|
||||
expect(result.heal).toBe(true);
|
||||
});
|
||||
|
||||
it('returns heal:false when healthStatus is not unhealthy', () => {
|
||||
const result = service.shouldHeal(
|
||||
{ ...baseState, healthStatus: 'healthy' },
|
||||
basePolicy,
|
||||
'container123',
|
||||
Date.now(),
|
||||
);
|
||||
expect(result.heal).toBe(false);
|
||||
});
|
||||
|
||||
it('returns heal:false when healthStatus is undefined', () => {
|
||||
const result = service.shouldHeal(
|
||||
{ ...baseState, healthStatus: undefined },
|
||||
basePolicy,
|
||||
'container123',
|
||||
Date.now(),
|
||||
);
|
||||
expect(result.heal).toBe(false);
|
||||
});
|
||||
|
||||
it('returns heal:false when state is undefined', () => {
|
||||
const result = service.shouldHeal(undefined, basePolicy, 'container123', Date.now());
|
||||
expect(result.heal).toBe(false);
|
||||
});
|
||||
|
||||
it('returns heal:false when unhealthySince is undefined', () => {
|
||||
const result = service.shouldHeal(
|
||||
{ ...baseState, unhealthySince: undefined },
|
||||
basePolicy,
|
||||
'container123',
|
||||
Date.now(),
|
||||
);
|
||||
expect(result.heal).toBe(false);
|
||||
});
|
||||
|
||||
it('returns heal:false when duration threshold is not yet met', () => {
|
||||
// Only 2 minutes, threshold is 5
|
||||
const state = { ...baseState, unhealthySince: Date.now() - 2 * 60_000 };
|
||||
const result = service.shouldHeal(state, basePolicy, 'container123', Date.now());
|
||||
expect(result.heal).toBe(false);
|
||||
expect(result.skipReason).toBe('duration_not_met');
|
||||
});
|
||||
|
||||
it('returns skipped_user_action when lastKillAt is within the window', () => {
|
||||
// 30s ago, well within the 60s INTENTIONAL_KILL_WINDOW_MS
|
||||
const state = { ...baseState, lastKillAt: Date.now() - 30_000 };
|
||||
const result = service.shouldHeal(state, basePolicy, 'container123', Date.now());
|
||||
expect(result.heal).toBe(false);
|
||||
expect(result.skipReason).toBe('skipped_user_action');
|
||||
});
|
||||
|
||||
it('does not suppress when lastKillAt is outside the intentional kill window', () => {
|
||||
const state = {
|
||||
...baseState,
|
||||
lastKillAt: Date.now() - (INTENTIONAL_KILL_WINDOW_MS + 5_000),
|
||||
};
|
||||
const result = service.shouldHeal(state, basePolicy, 'container123', Date.now());
|
||||
expect(result.heal).toBe(true);
|
||||
});
|
||||
|
||||
it('returns skipped_cooldown when last_fired_at is within cooldown period', () => {
|
||||
// Fired 5 min ago, cooldown is 10 min
|
||||
const policy = { ...basePolicy, last_fired_at: Date.now() - 5 * 60_000, cooldown_mins: 10 };
|
||||
const result = service.shouldHeal(baseState, policy, 'container123', Date.now());
|
||||
expect(result.heal).toBe(false);
|
||||
expect(result.skipReason).toBe('skipped_cooldown');
|
||||
});
|
||||
|
||||
it('does not apply cooldown when last_fired_at is 0', () => {
|
||||
const policy = { ...basePolicy, last_fired_at: 0 };
|
||||
const result = service.shouldHeal(baseState, policy, 'container123', Date.now());
|
||||
expect(result.heal).toBe(true);
|
||||
});
|
||||
|
||||
it('does not apply cooldown when last_fired_at exceeds the cooldown window', () => {
|
||||
// Fired 15 min ago, cooldown is 10 min
|
||||
const policy = { ...basePolicy, last_fired_at: Date.now() - 15 * 60_000, cooldown_mins: 10 };
|
||||
const result = service.shouldHeal(baseState, policy, 'container123', Date.now());
|
||||
expect(result.heal).toBe(true);
|
||||
});
|
||||
|
||||
it('returns skipped_rate_limit when hourly restart count is at the configured max', () => {
|
||||
const now = Date.now();
|
||||
// Pre-populate with 3 entries within the last hour (policy max is 3)
|
||||
const map = (service as any).restartTimestamps as Map<string, number[]>;
|
||||
map.set('container123', [now - 10_000, now - 20_000, now - 30_000]);
|
||||
const result = service.shouldHeal(baseState, basePolicy, 'container123', now);
|
||||
expect(result.heal).toBe(false);
|
||||
expect(result.skipReason).toBe('skipped_rate_limit');
|
||||
});
|
||||
|
||||
it('does not rate-limit when all timestamps are older than one hour', () => {
|
||||
const now = Date.now();
|
||||
const map = (service as any).restartTimestamps as Map<string, number[]>;
|
||||
// All entries are >1 hour old, so they fall outside the rate-limit window
|
||||
map.set('container123', [now - 70 * 60_000, now - 80 * 60_000, now - 90 * 60_000]);
|
||||
const result = service.shouldHeal(baseState, basePolicy, 'container123', now);
|
||||
expect(result.heal).toBe(true);
|
||||
});
|
||||
|
||||
it('counts only recent timestamps toward the rate limit', () => {
|
||||
const now = Date.now();
|
||||
const map = (service as any).restartTimestamps as Map<string, number[]>;
|
||||
// 2 old (outside window) + 1 recent = 1 active restart; max is 3, so still allowed
|
||||
map.set('container123', [now - 70 * 60_000, now - 80 * 60_000, now - 5_000]);
|
||||
const result = service.shouldHeal(baseState, basePolicy, 'container123', now);
|
||||
expect(result.heal).toBe(true);
|
||||
});
|
||||
|
||||
it('returns not_unhealthy as skipReason when container is healthy', () => {
|
||||
const result = service.shouldHeal(
|
||||
{ ...baseState, healthStatus: 'healthy' },
|
||||
basePolicy,
|
||||
'container123',
|
||||
Date.now(),
|
||||
);
|
||||
expect(result.skipReason).toBe('not_unhealthy');
|
||||
});
|
||||
});
|
||||
Reference in New Issue
Block a user