managarten/docker/prometheus/prometheus.yml
Till JS 0bf01f434e feat(mana-ai): Prometheus /metrics endpoint + status.mana.how integration
Wires mana-ai into the existing observability stack so tick throughput,
plan-failure rates, planner latencies, and snapshot refresh health are
visible in Grafana + Prometheus, and the service's uptime surfaces on
status.mana.how under the "Internal" section.

- `src/metrics.ts` — prom-client Registry with `mana_ai_` prefix.
  Counters: ticks_total, plans_produced_total, plans_written_back_total,
  parse_failures_total, mission_errors_total, snapshots_new/updated,
  snapshot_rows_applied_total, http_requests_total.
  Histograms: tick_duration_seconds (0.1–120s), planner_request_
  duration_seconds (0.25–60s), http_request_duration_seconds (0.005–10s).
- `src/index.ts` — HTTP middleware labels every request by
  method/path/status; `/metrics` serves the Prometheus text format.
- `src/cron/tick.ts` — increments counters + wraps the tick with
  `tickDuration.startTimer()`. Snapshot stats fold through.
- `src/planner/client.ts` — wraps `complete()` in a latency histogram
  timer so planner tail latency shows up separately from tick duration.
- `docker/prometheus/prometheus.yml` —
  1. New `mana-ai` scrape job against `mana-ai:3066/metrics` (30s).
  2. `/health` added to the `blackbox-internal` job so uptime shows on
     status.mana.how alongside mana-geocoding.
- `scripts/generate-status-page.sh` — friendly label for the new probe:
  `mana-ai:3066/health` → "Mana AI Runner" (generator already iterates
  `blackbox-internal`, no other changes needed).
- `package.json` — prom-client ^15.1.3

All 17 Bun tests still pass; tsc clean.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-04-15 01:41:40 +02:00

382 lines
11 KiB
YAML

# Mana Prometheus Configuration
# Scrapes metrics from all services
global:
scrape_interval: 15s
evaluation_interval: 15s
# Load alerting rules
rule_files:
- /etc/prometheus/alerts.yml
# Alertmanager configuration
alerting:
alertmanagers:
- static_configs:
- targets: ['alertmanager:9093']
scrape_configs:
# Prometheus self-monitoring
- job_name: 'prometheus'
static_configs:
- targets: ['localhost:9090']
# Host system metrics via node-exporter
- job_name: 'node'
static_configs:
- targets: ['node-exporter:9100']
relabel_configs:
- source_labels: [__address__]
target_label: instance
replacement: 'mac-mini'
# Docker container metrics via cAdvisor
- job_name: 'cadvisor'
static_configs:
- targets: ['cadvisor:8080']
# PostgreSQL metrics
- job_name: 'postgres'
static_configs:
- targets: ['postgres-exporter:9187']
# Redis metrics
- job_name: 'redis'
static_configs:
- targets: ['redis-exporter:9121']
# ============================================
# Core Services (Hono/Bun + Go)
# ============================================
# Auth Service
- job_name: 'mana-auth'
static_configs:
- targets: ['mana-auth:3001']
metrics_path: '/metrics'
scrape_interval: 30s
# Credits Service
- job_name: 'mana-credits'
static_configs:
- targets: ['mana-credits:3002']
metrics_path: '/metrics'
scrape_interval: 30s
# User Service
- job_name: 'mana-user'
static_configs:
- targets: ['mana-user:3062']
metrics_path: '/metrics'
scrape_interval: 30s
# Subscriptions Service
- job_name: 'mana-subscriptions'
static_configs:
- targets: ['mana-subscriptions:3063']
metrics_path: '/metrics'
scrape_interval: 30s
# Analytics Service
- job_name: 'mana-analytics'
static_configs:
- targets: ['mana-analytics:3064']
metrics_path: '/metrics'
scrape_interval: 30s
# ULoad Server
- job_name: 'uload-server'
static_configs:
- targets: ['mana-app-uload-server:3070']
metrics_path: '/metrics'
scrape_interval: 30s
# Memoro Server
- job_name: 'memoro-server'
static_configs:
- targets: ['mana-app-memoro-server:3015']
metrics_path: '/metrics'
scrape_interval: 30s
# NOTE: Individual app backends (chat, todo, calendar, contacts, storage,
# food, music, plants, picture) have been REMOVED — all migrated to
# local-first architecture. Only uload-server and memoro-server remain.
# Mana LLM Gateway (Ollama + Google Fallback)
- job_name: 'mana-llm'
static_configs:
- targets: ['mana-llm:3020']
metrics_path: '/metrics'
scrape_interval: 15s
# Mana Search Service
- job_name: 'mana-search'
static_configs:
- targets: ['mana-search:3012']
metrics_path: '/metrics'
scrape_interval: 30s
# Mana Media Service
- job_name: 'mana-media'
static_configs:
- targets: ['mana-media:3011']
metrics_path: '/metrics'
scrape_interval: 30s
# Mana AI Service (Bun) — background Mission Runner for the AI Workbench.
# Exposes tick stats, planner-request latencies, snapshot refresh
# counters, and standard HTTP metrics at /metrics.
- job_name: 'mana-ai'
static_configs:
- targets: ['mana-ai:3066']
metrics_path: '/metrics'
scrape_interval: 30s
# ============================================
# GPU Server (Windows PC, LAN: 192.168.178.11)
# ============================================
# GPU: LLM Gateway
- job_name: 'gpu-llm'
static_configs:
- targets: ['192.168.178.11:3025']
labels:
instance: 'gpu-server'
metrics_path: '/metrics'
scrape_interval: 15s
# GPU: Speech-to-Text (WhisperX)
- job_name: 'gpu-stt'
static_configs:
- targets: ['192.168.178.11:3020']
labels:
instance: 'gpu-server'
metrics_path: '/health'
scrape_interval: 30s
# GPU: Text-to-Speech
- job_name: 'gpu-tts'
static_configs:
- targets: ['192.168.178.11:3022']
labels:
instance: 'gpu-server'
metrics_path: '/health'
scrape_interval: 30s
# GPU: Image Generation (FLUX.2)
- job_name: 'gpu-image-gen'
static_configs:
- targets: ['192.168.178.11:3023']
labels:
instance: 'gpu-server'
metrics_path: '/health'
scrape_interval: 30s
# GPU: Video Generation (LTX-Video)
- job_name: 'gpu-video-gen'
static_configs:
- targets: ['192.168.178.11:3026']
labels:
instance: 'gpu-server'
metrics_path: '/health'
scrape_interval: 30s
# ============================================
# Go Infrastructure Services
# ============================================
# API Gateway (Go)
- job_name: 'mana-api-gateway'
static_configs:
- targets: ['mana-api-gateway:3016']
metrics_path: '/metrics'
scrape_interval: 15s
# Sync Server (Go) — local-first data sync
- job_name: 'mana-sync'
static_configs:
- targets: ['mana-core-sync:3051']
metrics_path: '/metrics'
scrape_interval: 30s
# Notification Service (Go) — email, push, webhook
- job_name: 'mana-notify'
static_configs:
- targets: ['mana-core-notify:3013']
metrics_path: '/metrics'
scrape_interval: 30s
# Crawler Service (Go)
- job_name: 'mana-crawler'
static_configs:
- targets: ['mana-crawler:3014']
metrics_path: '/metrics'
scrape_interval: 30s
# ============================================
# Blackbox Exporter — HTTP Uptime Probes
# ============================================
# Web Apps (Unified Mana app at mana.how + standalone games)
- job_name: 'blackbox-web'
metrics_path: /probe
params:
module: [http_2xx]
static_configs:
- targets:
# Unified Mana app (all modules as routes)
- https://mana.how
- https://mana.how/chat
- https://mana.how/todo
- https://mana.how/calendar
- https://mana.how/contacts
- https://mana.how/times
- https://mana.how/photos
- https://mana.how/picture
- https://mana.how/storage
- https://mana.how/presi
- https://mana.how/food
- https://mana.how/plants
- https://mana.how/calc
- https://mana.how/quotes
- https://mana.how/cards
- https://mana.how/skilltree
- https://mana.how/music
- https://mana.how/citycorners
- https://mana.how/memoro
- https://mana.how/moodlit
- https://mana.how/context
- https://mana.how/questions
- https://mana.how/uload
- https://mana.how/notes
- https://mana.how/habits
- https://mana.how/guides
- https://mana.how/inventory
- https://mana.how/body
- https://mana.how/journal
- https://mana.how/dreams
- https://mana.how/firsts
- https://mana.how/period
- https://mana.how/events
- https://mana.how/finance
- https://mana.how/places
- https://mana.how/who
- https://mana.how/news
- https://mana.how/mail
- https://mana.how/playground
# Standalone games (separate containers)
- https://whopxl.mana.how
- https://arcade.mana.how
relabel_configs:
- source_labels: [__address__]
target_label: __param_target
- source_labels: [__param_target]
target_label: instance
- target_label: __address__
replacement: blackbox-exporter:9115
# API Health Endpoints (only services with running containers)
- job_name: 'blackbox-api'
metrics_path: /probe
params:
module: [http_health]
static_configs:
- targets:
- https://auth.mana.how/health
- https://api.mana.how/health
relabel_configs:
- source_labels: [__address__]
target_label: __param_target
- source_labels: [__param_target]
target_label: instance
- target_label: __address__
replacement: blackbox-exporter:9115
# Internal-only services (not exposed via Cloudflare).
# Probed over the Docker network so the blackbox exporter reaches
# them by container name.
- job_name: 'blackbox-internal'
metrics_path: /probe
params:
module: [http_2xx]
static_configs:
- targets:
# mana-geocoding's own health (Hono wrapper)
- http://mana-geocoding:3018/health
# Upstream Pelias health, proxied through the wrapper so the
# blackbox-exporter doesn't need host.docker.internal access.
- http://mana-geocoding:3018/health/pelias
# mana-ai (Mission Runner) — internal-only, no CF tunnel.
- http://mana-ai:3066/health
relabel_configs:
- source_labels: [__address__]
target_label: __param_target
- source_labels: [__param_target]
target_label: instance
- target_label: __address__
replacement: blackbox-exporter:9115
# Infrastructure & Monitoring Tools
- job_name: 'blackbox-infra'
metrics_path: /probe
params:
module: [http_2xx]
static_configs:
- targets:
- https://git.mana.how
- https://grafana.mana.how
- https://stats.mana.how
- https://glitchtip.mana.how
relabel_configs:
- source_labels: [__address__]
target_label: __param_target
- source_labels: [__param_target]
target_label: instance
- target_label: __address__
replacement: blackbox-exporter:9115
# GPU Server Services — probe /health, not /
# The GPU services (whisper STT, TTS, FLUX image gen) only return 2xx
# on /health; their root path returns 401/403/404 by design (auth or
# API-only). Ollama is the exception — its / returns 200, but it has
# no /health endpoint, so we keep it on / via a separate target.
- job_name: 'blackbox-gpu'
metrics_path: /probe
params:
module: [http_health]
static_configs:
- targets:
- https://gpu-stt.mana.how/health
- https://gpu-tts.mana.how/health
- https://gpu-img.mana.how/health
- https://gpu-video.mana.how/health
relabel_configs:
- source_labels: [__address__]
target_label: __param_target
- source_labels: [__param_target]
target_label: instance
- target_label: __address__
replacement: blackbox-exporter:9115
- job_name: 'blackbox-gpu-root'
metrics_path: /probe
params:
module: [http_2xx]
static_configs:
- targets:
- https://gpu-ollama.mana.how
relabel_configs:
- source_labels: [__address__]
target_label: __param_target
- source_labels: [__param_target]
target_label: instance
- target_label: __address__
replacement: blackbox-exporter:9115
# ============================================
# Pushgateway (deploy metrics, batch jobs)
# ============================================
- job_name: 'pushgateway'
honor_labels: true
static_configs:
- targets: ['pushgateway:9091']