feat(health): add model and provider health monitoring

- Rename the original health monitor to endpoint health monitor - Add tab navigation for endpoint, model, and provider health views - Add model health monitor cards with availability, latency, first-byte latency, and 60-point history - Add admin-only provider health monitor with collapsible active-provider sections - Show per-provider model health cards after expanding a provider - Add backend model health and provider health monitor payload builders - Add admin endpoint for provider health monitoring - Add provider-scoped usage breakdown filtering for per-provider model statistics - Add frontend API types and request helpers for model/provider health data - Add demo mock data for model and provider health monitoring - Fix model health timeline time-unit handling so request history segments render correctly Verification: - cargo fmt - npm run type-check - npm run build - cargo test -p aether-gateway health_models - cargo test -p aether-gateway health_providers - cargo test -p aether-gateway gateway_exposes_frontdoor_manifest_without_proxying_upstream
2026-10-04 16:37:46 +08:00 · 2026-05-28 12:00:20 +08:00
parent 14ad6e9b75
commit 4ce056fe45
24 changed files with 1914 additions and 17 deletions
@@ -139,6 +139,20 @@ function generateHealthEvents(
  return events.sort((a, b) => new Date(a.timestamp).getTime() - new Date(b.timestamp).getTime())
 }

+function generateHealthTimeline(
+  healthyRate: number,
+  warningRate: number,
+  segments = 60
+) {
+  return Array.from({ length: segments }, () => {
+    const rand = Math.random()
+    if (rand < healthyRate) return 'healthy'
+    if (rand < healthyRate + warningRate) return 'warning'
+    if (rand < 0.96) return 'unknown'
+    return 'unhealthy'
+  })
+}
+
 // Mock 端点健康数据
 // 注意：success_rate 使用 0-1 之间的小数，前端会乘以 100 显示为百分比
 // 事件的成功/失败/跳过比例必须与 success_rate 保持一致
@@ -246,6 +260,154 @@ const MOCK_ENDPOINT_STATUS = {
  ]
 }

+const MOCK_MODEL_STATUS = {
+  generated_at: new Date().toISOString(),
+  models: [
+    {
+      model: 'gpt-5.5',
+      display_name: 'gpt-5.5',
+      total_attempts: 2021,
+      success_count: 2000,
+      failed_count: 21,
+      success_rate: 0.9896,
+      avg_latency_ms: 1736,
+      avg_first_byte_ms: 176,
+      provider_count: 3,
+      last_event_at: new Date().toISOString(),
+      events: generateHealthEvents(60, 0.989, 0.008, 0.003, 1600, 460),
+      timeline: generateHealthTimeline(0.9, 0.05),
+      time_range_start: new Date(Date.now() - 6 * 60 * 60 * 1000).toISOString(),
+      time_range_end: new Date().toISOString()
+    },
+    {
+      model: 'claude-sonnet-4-5-20250929',
+      display_name: 'Claude Sonnet 4.5',
+      total_attempts: 1684,
+      success_count: 1642,
+      failed_count: 42,
+      success_rate: 0.9751,
+      avg_latency_ms: 1280,
+      avg_first_byte_ms: 221,
+      provider_count: 2,
+      last_event_at: new Date().toISOString(),
+      events: generateHealthEvents(60, 0.975, 0.02, 0.005, 1200, 520),
+      timeline: generateHealthTimeline(0.84, 0.09),
+      time_range_start: new Date(Date.now() - 6 * 60 * 60 * 1000).toISOString(),
+      time_range_end: new Date().toISOString()
+    },
+    {
+      model: 'gemini-3-pro-preview',
+      display_name: 'Gemini 3 Pro Preview',
+      total_attempts: 932,
+      success_count: 887,
+      failed_count: 45,
+      success_rate: 0.9517,
+      avg_latency_ms: 940,
+      avg_first_byte_ms: 184,
+      provider_count: 2,
+      last_event_at: new Date().toISOString(),
+      events: generateHealthEvents(55, 0.952, 0.04, 0.008, 860, 300),
+      timeline: generateHealthTimeline(0.78, 0.14),
+      time_range_start: new Date(Date.now() - 6 * 60 * 60 * 1000).toISOString(),
+      time_range_end: new Date().toISOString()
+    },
+    {
+      model: 'gpt-5.1-codex-mini',
+      display_name: 'gpt-5.1-codex-mini',
+      total_attempts: 418,
+      success_count: 349,
+      failed_count: 69,
+      success_rate: 0.835,
+      avg_latency_ms: 2310,
+      avg_first_byte_ms: 420,
+      provider_count: 1,
+      last_event_at: new Date().toISOString(),
+      events: generateHealthEvents(45, 0.835, 0.145, 0.02, 2200, 780),
+      timeline: generateHealthTimeline(0.58, 0.24),
+      time_range_start: new Date(Date.now() - 6 * 60 * 60 * 1000).toISOString(),
+      time_range_end: new Date().toISOString()
+    }
+  ]
+}
+
+const MOCK_PROVIDER_HEALTH_STATUS = {
+  generated_at: new Date().toISOString(),
+  providers: [
+    {
+      provider_id: 'provider-001',
+      provider_name: 'OpenAI Official',
+      provider_type: 'codex',
+      is_active: true,
+      total_attempts: 2021,
+      success_count: 2000,
+      failed_count: 21,
+      success_rate: 0.9896,
+      avg_latency_ms: 1736,
+      avg_first_byte_ms: 176,
+      model_count: 2,
+      last_event_at: new Date().toISOString(),
+      timeline: generateHealthTimeline(0.9, 0.05),
+      time_range_start: new Date(Date.now() - 6 * 60 * 60 * 1000).toISOString(),
+      time_range_end: new Date().toISOString(),
+      models: [MOCK_MODEL_STATUS.models[0], MOCK_MODEL_STATUS.models[3]]
+    },
+    {
+      provider_id: 'provider-002',
+      provider_name: 'Anthropic Official',
+      provider_type: 'claude_code',
+      is_active: true,
+      total_attempts: 1684,
+      success_count: 1642,
+      failed_count: 42,
+      success_rate: 0.9751,
+      avg_latency_ms: 1280,
+      avg_first_byte_ms: 221,
+      model_count: 1,
+      last_event_at: new Date().toISOString(),
+      timeline: generateHealthTimeline(0.84, 0.09),
+      time_range_start: new Date(Date.now() - 6 * 60 * 60 * 1000).toISOString(),
+      time_range_end: new Date().toISOString(),
+      models: [MOCK_MODEL_STATUS.models[1]]
+    },
+    {
+      provider_id: 'provider-003',
+      provider_name: 'Google AI',
+      provider_type: 'gemini_cli',
+      is_active: true,
+      total_attempts: 932,
+      success_count: 887,
+      failed_count: 45,
+      success_rate: 0.9517,
+      avg_latency_ms: 940,
+      avg_first_byte_ms: 184,
+      model_count: 1,
+      last_event_at: new Date().toISOString(),
+      timeline: generateHealthTimeline(0.78, 0.14),
+      time_range_start: new Date(Date.now() - 6 * 60 * 60 * 1000).toISOString(),
+      time_range_end: new Date().toISOString(),
+      models: [MOCK_MODEL_STATUS.models[2]]
+    },
+    {
+      provider_id: 'provider-004',
+      provider_name: 'AWS Bedrock',
+      provider_type: 'custom',
+      is_active: true,
+      total_attempts: 0,
+      success_count: 0,
+      failed_count: 0,
+      success_rate: 1,
+      avg_latency_ms: null,
+      avg_first_byte_ms: null,
+      model_count: 0,
+      last_event_at: null,
+      timeline: Array.from({ length: 60 }, () => 'unknown'),
+      time_range_start: new Date(Date.now() - 6 * 60 * 60 * 1000).toISOString(),
+      time_range_end: new Date().toISOString(),
+      models: []
+    }
+  ]
+}
+
 // 生成活跃热力图数据（最近365天）
 function generateActivityHeatmap() {
  const days: Array<{
@@ -1026,6 +1188,18 @@ const mockHandlers: Record<string, (config: AxiosRequestConfig) => Promise<Axios
    return createMockResponse(MOCK_ENDPOINT_STATUS)
  },

+  'GET /api/admin/endpoints/health/models': async () => {
+    await delay()
+    requireAdmin()
+    return createMockResponse(MOCK_MODEL_STATUS)
+  },
+
+  'GET /api/admin/endpoints/health/providers': async () => {
+    await delay()
+    requireAdmin()
+    return createMockResponse(MOCK_PROVIDER_HEALTH_STATUS)
+  },
+
  'GET /api/admin/endpoints/keys': async () => {
    await delay()
    requireAdmin()
@@ -1387,6 +1561,28 @@ const mockHandlers: Record<string, (config: AxiosRequestConfig) => Promise<Axios
        events: f.events.slice(0, 10)
      }))
    })
+  },
+
+  'GET /api/public/health/models': async () => {
+    await delay()
+    return createMockResponse({
+      generated_at: new Date().toISOString(),
+      models: MOCK_MODEL_STATUS.models.map(model => ({
+        model: model.model,
+        display_name: model.display_name,
+        total_attempts: model.total_attempts,
+        success_count: model.success_count,
+        failed_count: model.failed_count,
+        success_rate: model.success_rate,
+        avg_latency_ms: model.avg_latency_ms,
+        avg_first_byte_ms: model.avg_first_byte_ms,
+        last_event_at: model.last_event_at,
+        events: model.events.slice(0, 10),
+        timeline: model.timeline,
+        time_range_start: model.time_range_start,
+        time_range_end: model.time_range_end
+      }))
+    })
  }
 }