Add external monitoring stack and alerting docs

This commit is contained in:
2569718930@qq.com
2026-04-01 01:49:58 +08:00
parent 6626b5472c
commit a8462e188b
13 changed files with 729 additions and 1 deletions
+7
View File
@@ -18,6 +18,12 @@ OPEN_METEO_DISK_CACHE_PATH=/var/lib/polyweather/open_meteo_cache.json
UID=1000
GID=1000
POLYWEATHER_STATE_STORAGE_MODE=dual
POLYWEATHER_PROMETHEUS_PORT=9090
POLYWEATHER_ALERTMANAGER_PORT=9093
POLYWEATHER_ALERT_RELAY_PORT=9099
POLYWEATHER_GRAFANA_PORT=3001
POLYWEATHER_GRAFANA_ADMIN_USER=admin
POLYWEATHER_GRAFANA_ADMIN_PASSWORD=polyweather
########################################
# 2) Telegram bot minimal
@@ -72,6 +78,7 @@ TELEGRAM_ALERT_MIN_TRIGGER_COUNT=2
TELEGRAM_ALERT_MIN_SEVERITY=medium
TELEGRAM_ALERT_MISPRICING_MAX_YES_BUY=0.10
TELEGRAM_ALERT_CITIES=ankara,london,paris,seoul,hong kong,shanghai,singapore,tokyo,tel aviv,toronto,buenos aires,wellington,new york,chicago,dallas,miami,atlanta,seattle,lucknow,sao paulo,munich
POLYWEATHER_MONITORING_ALERT_CHAT_IDS=
########################################
# 6) Frontend-facing shared values
+18
View File
@@ -23,6 +23,7 @@
- 已上线支付运行态与审计接口:`/api/payments/runtime`
- 已上线轻量运营后台:`/ops`(会员、周榜、补分、支付异常单)。
- 已上线轻量可观测性:`/healthz``/api/system/status``/metrics`
- 已补最小外部监控栈:Prometheus + Alertmanager + Grafana + Telegram 告警 relay。
- 运行态状态与缓存已支持 SQLite 渐进迁移:`file / dual / sqlite`
- 已接入 EMOS/CRPS 校准链路,但当前仍保持 `emos_shadow`
@@ -122,6 +123,22 @@ curl http://127.0.0.1:8000/metrics
docker compose logs -f polyweather | egrep "payment event loop started|payment confirm loop started|payment auto-confirmed"
```
### 外部监控栈
```bash
docker compose --profile monitoring up -d polyweather_prometheus polyweather_alertmanager polyweather_alert_relay polyweather_grafana
```
- Prometheus`http://127.0.0.1:${POLYWEATHER_PROMETHEUS_PORT:-9090}`
- Alertmanager`http://127.0.0.1:${POLYWEATHER_ALERTMANAGER_PORT:-9093}`
- Grafana`http://127.0.0.1:${POLYWEATHER_GRAFANA_PORT:-3001}`
手动巡检:
```bash
python scripts/check_ops_health.py --base-url http://127.0.0.1:8000
```
### 支付运行态
```bash
@@ -171,6 +188,7 @@ docker compose logs -f polyweather | egrep "polymarket wallet activity watcher s
- 支付审计说明:[docs/payments/PAYMENT_AUDIT_ZH.md](docs/payments/PAYMENT_AUDIT_ZH.md)
- 支付 V2 升级方案:[docs/payments/PAYMENT_UPGRADE_V2_ZH.md](docs/payments/PAYMENT_UPGRADE_V2_ZH.md)
- 运营后台说明:[docs/OPS_ADMIN_ZH.md](docs/OPS_ADMIN_ZH.md)
- 外部监控说明:[docs/MONITORING_ZH.md](docs/MONITORING_ZH.md)
- 深度评估报告:[docs/deep-research-report.md](docs/deep-research-report.md)
- 前端报告:[FRONTEND_REDESIGN_REPORT.md](FRONTEND_REDESIGN_REPORT.md)
- 发布流程:[RELEASE.md](RELEASE.md)
+66
View File
@@ -33,3 +33,69 @@ services:
- "8000:8000"
# UID/GID are mainly useful on Linux hosts to avoid root-owned output files.
user: "${UID:-1000}:${GID:-1000}"
polyweather_prometheus:
image: prom/prometheus:v3.4.1
container_name: polyweather_prometheus
restart: unless-stopped
profiles: ["monitoring"]
depends_on:
- polyweather_web
command:
- "--config.file=/etc/prometheus/prometheus.yml"
- "--storage.tsdb.path=/prometheus"
- "--storage.tsdb.retention.time=15d"
- "--web.enable-lifecycle"
volumes:
- ./monitoring/prometheus/prometheus.yml:/etc/prometheus/prometheus.yml:ro
- ./monitoring/prometheus/alerts.yml:/etc/prometheus/alerts.yml:ro
- ${POLYWEATHER_RUNTIME_DATA_DIR:-/var/lib/polyweather}/monitoring/prometheus:/prometheus
ports:
- "${POLYWEATHER_PROMETHEUS_PORT:-9090}:9090"
polyweather_alertmanager:
image: prom/alertmanager:v0.28.1
container_name: polyweather_alertmanager
restart: unless-stopped
profiles: ["monitoring"]
depends_on:
- polyweather_alert_relay
command:
- "--config.file=/etc/alertmanager/alertmanager.yml"
- "--storage.path=/alertmanager"
volumes:
- ./monitoring/alertmanager/alertmanager.yml:/etc/alertmanager/alertmanager.yml:ro
- ${POLYWEATHER_RUNTIME_DATA_DIR:-/var/lib/polyweather}/monitoring/alertmanager:/alertmanager
ports:
- "${POLYWEATHER_ALERTMANAGER_PORT:-9093}:9093"
polyweather_alert_relay:
<<: *polyweather-base
container_name: polyweather_alert_relay
restart: unless-stopped
profiles: ["monitoring"]
command: python scripts/alertmanager_telegram_relay.py
volumes:
- ${POLYWEATHER_RUNTIME_DATA_DIR:-/var/lib/polyweather}:/var/lib/polyweather
- ${POLYWEATHER_RUNTIME_DATA_DIR:-/var/lib/polyweather}:/app/data
ports:
- "${POLYWEATHER_ALERT_RELAY_PORT:-9099}:9099"
user: "${UID:-1000}:${GID:-1000}"
polyweather_grafana:
image: grafana/grafana-oss:12.0.2
container_name: polyweather_grafana
restart: unless-stopped
profiles: ["monitoring"]
depends_on:
- polyweather_prometheus
environment:
GF_SECURITY_ADMIN_USER: ${POLYWEATHER_GRAFANA_ADMIN_USER:-admin}
GF_SECURITY_ADMIN_PASSWORD: ${POLYWEATHER_GRAFANA_ADMIN_PASSWORD:-polyweather}
GF_USERS_ALLOW_SIGN_UP: "false"
volumes:
- ./monitoring/grafana/provisioning:/etc/grafana/provisioning:ro
- ./monitoring/grafana/dashboards:/var/lib/grafana/dashboards:ro
- ${POLYWEATHER_RUNTIME_DATA_DIR:-/var/lib/polyweather}/monitoring/grafana:/var/lib/grafana
ports:
- "${POLYWEATHER_GRAFANA_PORT:-3001}:3000"
+123
View File
@@ -0,0 +1,123 @@
# 外部监控与告警说明
最后更新:`2026-04-01`
## 1. 目标
在现有轻量可观测性基础上,把 PolyWeather 补成最小可用的外部监控链路:
- Prometheus 抓取 `/metrics`
- Alertmanager 根据规则聚合告警
- Telegram relay 把告警推到运营频道
- Grafana 展示趋势面板
- 巡检脚本补健康检查
## 2. 组件
本仓库现在内置 4 个监控组件:
- `polyweather_prometheus`
- `polyweather_alertmanager`
- `polyweather_alert_relay`
- `polyweather_grafana`
对应配置目录:
- [monitoring/prometheus/prometheus.yml](../monitoring/prometheus/prometheus.yml)
- [monitoring/prometheus/alerts.yml](../monitoring/prometheus/alerts.yml)
- [monitoring/alertmanager/alertmanager.yml](../monitoring/alertmanager/alertmanager.yml)
- [monitoring/grafana/dashboards/polyweather-overview.json](../monitoring/grafana/dashboards/polyweather-overview.json)
## 3. 启动
```bash
docker compose --profile monitoring up -d polyweather_prometheus polyweather_alertmanager polyweather_alert_relay polyweather_grafana
```
默认端口:
- Prometheus: `9090`
- Alertmanager: `9093`
- Grafana: `3001`
- Alert relay: `9099`
## 4. 环境变量
在 [.env.example](../.env.example) 里新增了这些配置:
```env
POLYWEATHER_PROMETHEUS_PORT=9090
POLYWEATHER_ALERTMANAGER_PORT=9093
POLYWEATHER_ALERT_RELAY_PORT=9099
POLYWEATHER_GRAFANA_PORT=3001
POLYWEATHER_GRAFANA_ADMIN_USER=admin
POLYWEATHER_GRAFANA_ADMIN_PASSWORD=polyweather
POLYWEATHER_MONITORING_ALERT_CHAT_IDS=
```
说明:
- `POLYWEATHER_MONITORING_ALERT_CHAT_IDS` 为空时,relay 会自动回退到:
- `TELEGRAM_CHAT_IDS`
- `TELEGRAM_CHAT_ID`
- 告警发送仍复用现有 `TELEGRAM_BOT_TOKEN`
## 5. 当前告警规则
当前默认规则:
- `PolyWeatherWebDown`
- `PolyWeatherHttp5xxBurst`
- `PolyWeatherHighSourceErrorRate`
- `PolyWeatherOpenMeteoCooldownLoop`
- `PolyWeatherSlowHttpAverage`
规则文件:
- [monitoring/prometheus/alerts.yml](../monitoring/prometheus/alerts.yml)
## 6. 当前 Grafana 面板
预置了一个最小仪表板:
- `PolyWeather Overview`
包含这些图:
- HTTP Requests by Status
- HTTP Latency
- Source Requests by Outcome
- Source Error Rate (15m)
## 7. 巡检脚本
手动巡检:
```bash
python scripts/check_ops_health.py --base-url http://127.0.0.1:8000
```
这个脚本会检查:
- `/healthz`
- `/api/system/status`
- `/metrics`
任何一项失败都会非零退出,适合挂到 crontab 或 systemd timer。
## 8. 备注
这套监控现在已经具备:
- 外部抓取
- 告警规则
- Telegram 推送
- 趋势面板
- 巡检脚本
但它仍是“最小可用版”,还没有覆盖:
- 节点级 CPU / 内存 / 磁盘
- 数据库体积趋势
- 更细粒度支付指标
- 按城市/来源拆分的业务 SLA
+6 -1
View File
@@ -1,6 +1,6 @@
# Ops 运营后台说明
最后更新:`2026-03-21`
最后更新:`2026-04-01`
## 1. 入口
@@ -28,6 +28,7 @@ POLYWEATHER_OPS_ADMIN_EMAILS=yhrsc30@gmail.com
- 当前会员
- 周榜
- 支付异常单
- 漏斗转化面板
### 写能力
@@ -102,3 +103,7 @@ python scripts/reconcile_subscription_by_email.py --email <user_email>
- 让会员、积分、支付事故、系统状态可查
- 让常见人工操作不必再直接写 SQL
外部监控与告警栈说明见:
- [MONITORING_ZH.md](./MONITORING_ZH.md)
+15
View File
@@ -0,0 +1,15 @@
global:
resolve_timeout: 5m
route:
receiver: telegram-relay
group_by: ["alertname"]
group_wait: 30s
group_interval: 5m
repeat_interval: 3h
receivers:
- name: telegram-relay
webhook_configs:
- url: http://polyweather_alert_relay:9099/alerts
send_resolved: true
@@ -0,0 +1,227 @@
{
"annotations": {
"list": [
{
"builtIn": 1,
"datasource": {
"type": "grafana",
"uid": "-- Grafana --"
},
"enable": true,
"hide": true,
"iconColor": "rgba(0, 211, 255, 1)",
"name": "Annotations & Alerts",
"type": "dashboard"
}
]
},
"editable": true,
"fiscalYearStartMonth": 0,
"graphTooltip": 0,
"id": null,
"links": [],
"panels": [
{
"datasource": {
"type": "prometheus",
"uid": "prometheus"
},
"fieldConfig": {
"defaults": {
"color": {
"mode": "palette-classic"
},
"unit": "reqps"
},
"overrides": []
},
"gridPos": {
"h": 8,
"w": 12,
"x": 0,
"y": 0
},
"id": 1,
"options": {
"legend": {
"displayMode": "table",
"placement": "bottom",
"showLegend": true
},
"tooltip": {
"mode": "single"
}
},
"targets": [
{
"expr": "sum by (status) (rate(polyweather_http_requests_total[5m]))",
"legendFormat": "{{status}}",
"refId": "A"
}
],
"title": "HTTP Requests by Status",
"type": "timeseries"
},
{
"datasource": {
"type": "prometheus",
"uid": "prometheus"
},
"fieldConfig": {
"defaults": {
"color": {
"mode": "palette-classic"
},
"unit": "ms"
},
"overrides": []
},
"gridPos": {
"h": 8,
"w": 12,
"x": 12,
"y": 0
},
"id": 2,
"options": {
"legend": {
"displayMode": "table",
"placement": "bottom",
"showLegend": true
},
"tooltip": {
"mode": "single"
}
},
"targets": [
{
"expr": "sum(rate(polyweather_http_request_duration_ms_sum[5m])) / clamp_min(sum(rate(polyweather_http_request_duration_ms_count[5m])), 1)",
"legendFormat": "avg",
"refId": "A"
},
{
"expr": "max(polyweather_http_request_duration_ms_max)",
"legendFormat": "max",
"refId": "B"
}
],
"title": "HTTP Latency",
"type": "timeseries"
},
{
"datasource": {
"type": "prometheus",
"uid": "prometheus"
},
"fieldConfig": {
"defaults": {
"color": {
"mode": "palette-classic"
},
"unit": "reqps"
},
"overrides": []
},
"gridPos": {
"h": 8,
"w": 12,
"x": 0,
"y": 8
},
"id": 3,
"options": {
"legend": {
"displayMode": "table",
"placement": "bottom",
"showLegend": true
},
"tooltip": {
"mode": "single"
}
},
"targets": [
{
"expr": "sum by (source, outcome) (rate(polyweather_source_requests_total[5m]))",
"legendFormat": "{{source}} / {{outcome}}",
"refId": "A"
}
],
"title": "Source Requests by Outcome",
"type": "timeseries"
},
{
"datasource": {
"type": "prometheus",
"uid": "prometheus"
},
"fieldConfig": {
"defaults": {
"color": {
"mode": "thresholds"
},
"thresholds": {
"mode": "absolute",
"steps": [
{
"color": "green"
},
{
"color": "orange",
"value": 10
},
{
"color": "red",
"value": 25
}
]
},
"unit": "percentunit"
},
"overrides": []
},
"gridPos": {
"h": 8,
"w": 12,
"x": 12,
"y": 8
},
"id": 4,
"options": {
"reduceOptions": {
"calcs": [
"lastNotNull"
],
"fields": "",
"values": false
}
},
"targets": [
{
"expr": "(sum(increase(polyweather_source_requests_total{outcome!~\"success|cache_hit\"}[15m])) / clamp_min(sum(increase(polyweather_source_requests_total[15m])), 1))",
"refId": "A"
}
],
"title": "Source Error Rate (15m)",
"type": "stat"
}
],
"refresh": "30s",
"schemaVersion": 41,
"style": "dark",
"tags": [
"polyweather",
"ops"
],
"templating": {
"list": []
},
"time": {
"from": "now-6h",
"to": "now"
},
"timepicker": {},
"timezone": "browser",
"title": "PolyWeather Overview",
"uid": "polyweather-overview",
"version": 1
}
@@ -0,0 +1,11 @@
apiVersion: 1
providers:
- name: PolyWeather Dashboards
orgId: 1
folder: PolyWeather
type: file
disableDeletion: false
editable: true
options:
path: /var/lib/grafana/dashboards
@@ -0,0 +1,9 @@
apiVersion: 1
datasources:
- name: PolyWeather Prometheus
type: prometheus
access: proxy
url: http://polyweather_prometheus:9090
isDefault: true
editable: false
+58
View File
@@ -0,0 +1,58 @@
groups:
- name: polyweather-runtime
rules:
- alert: PolyWeatherWebDown
expr: up{job="polyweather-web"} == 0
for: 2m
labels:
severity: critical
annotations:
summary: "PolyWeather web metrics endpoint is down"
description: "/metrics on polyweather_web has been unreachable for more than 2 minutes."
- alert: PolyWeatherHttp5xxBurst
expr: sum(increase(polyweather_http_requests_total{status=~"5.."}[10m])) > 10
for: 5m
labels:
severity: warning
annotations:
summary: "PolyWeather HTTP 5xx burst"
description: "More than 10 server errors were observed in the last 10 minutes."
- alert: PolyWeatherHighSourceErrorRate
expr: |
(
sum(increase(polyweather_source_requests_total{outcome!~"success|cache_hit"}[15m]))
/
clamp_min(sum(increase(polyweather_source_requests_total[15m])), 1)
) > 0.25
and sum(increase(polyweather_source_requests_total[15m])) > 20
for: 10m
labels:
severity: warning
annotations:
summary: "PolyWeather source error rate is high"
description: "External weather source errors exceeded 25% over the last 15 minutes."
- alert: PolyWeatherOpenMeteoCooldownLoop
expr: sum(increase(polyweather_source_requests_total{source="open_meteo",outcome=~"cooldown_skip|error"}[15m])) > 20
for: 10m
labels:
severity: warning
annotations:
summary: "Open-Meteo is rate-limited or erroring"
description: "Open-Meteo has produced repeated cooldown skips or errors in the last 15 minutes."
- alert: PolyWeatherSlowHttpAverage
expr: |
(
sum(rate(polyweather_http_request_duration_ms_sum[10m]))
/
clamp_min(sum(rate(polyweather_http_request_duration_ms_count[10m])), 1)
) > 2000
for: 10m
labels:
severity: warning
annotations:
summary: "PolyWeather average HTTP latency is high"
description: "Average HTTP request duration exceeded 2s over the last 10 minutes."
+19
View File
@@ -0,0 +1,19 @@
global:
scrape_interval: 30s
evaluation_interval: 30s
rule_files:
- /etc/prometheus/alerts.yml
alerting:
alertmanagers:
- static_configs:
- targets:
- polyweather_alertmanager:9093
scrape_configs:
- job_name: polyweather-web
metrics_path: /metrics
static_configs:
- targets:
- polyweather_web:8000
+103
View File
@@ -0,0 +1,103 @@
from __future__ import annotations
import json
import os
from http.server import BaseHTTPRequestHandler, HTTPServer
from typing import Any, Dict, List
import requests
def _chat_ids() -> List[str]:
raw = (
os.getenv("POLYWEATHER_MONITORING_ALERT_CHAT_IDS")
or os.getenv("TELEGRAM_CHAT_IDS")
or os.getenv("TELEGRAM_CHAT_ID")
or ""
)
return [item.strip() for item in raw.split(",") if item.strip()]
def _format_alerts(payload: Dict[str, Any]) -> str:
alerts = payload.get("alerts") or []
if not isinstance(alerts, list) or not alerts:
return "PolyWeather monitoring received an empty alert payload."
lines = ["PolyWeather monitoring alert"]
for alert in alerts[:10]:
if not isinstance(alert, dict):
continue
status = str(alert.get("status") or "unknown").upper()
labels = alert.get("labels") or {}
annotations = alert.get("annotations") or {}
alert_name = labels.get("alertname") or "unknown_alert"
severity = labels.get("severity") or "info"
summary = annotations.get("summary") or annotations.get("description") or ""
lines.append(f"- [{status}] {alert_name} ({severity})")
if summary:
lines.append(f" {summary}")
return "\n".join(lines)
def _send_telegram_message(text: str) -> None:
token = str(os.getenv("TELEGRAM_BOT_TOKEN") or "").strip()
chat_ids = _chat_ids()
if not token or not chat_ids:
return
url = f"https://api.telegram.org/bot{token}/sendMessage"
for chat_id in chat_ids:
try:
requests.post(
url,
json={
"chat_id": chat_id,
"text": text,
"disable_web_page_preview": True,
},
timeout=10,
).raise_for_status()
except Exception:
continue
class _Handler(BaseHTTPRequestHandler):
def do_GET(self) -> None: # noqa: N802
if self.path.rstrip("/") == "/healthz":
self.send_response(200)
self.send_header("Content-Type", "application/json; charset=utf-8")
self.end_headers()
self.wfile.write(b'{"ok":true}')
return
self.send_error(404)
def do_POST(self) -> None: # noqa: N802
if self.path.rstrip("/") != "/alerts":
self.send_error(404)
return
length = int(self.headers.get("Content-Length", "0") or "0")
raw = self.rfile.read(length) if length > 0 else b"{}"
try:
payload = json.loads(raw.decode("utf-8"))
except Exception:
self.send_error(400, "invalid json")
return
if not isinstance(payload, dict):
self.send_error(400, "invalid payload")
return
_send_telegram_message(_format_alerts(payload))
self.send_response(200)
self.send_header("Content-Type", "application/json; charset=utf-8")
self.end_headers()
self.wfile.write(b'{"ok":true}')
def log_message(self, format: str, *args: object) -> None: # noqa: A003
return
def main() -> None:
port = 9099
server = HTTPServer(("0.0.0.0", port), _Handler)
server.serve_forever()
if __name__ == "__main__":
main()
+67
View File
@@ -0,0 +1,67 @@
from __future__ import annotations
import argparse
import json
import sys
from typing import Dict, Tuple
import requests
def _get_json(url: str, timeout: float) -> Tuple[int, Dict]:
response = requests.get(url, timeout=timeout)
response.raise_for_status()
return response.status_code, response.json()
def _get_text(url: str, timeout: float) -> Tuple[int, str]:
response = requests.get(url, timeout=timeout)
response.raise_for_status()
return response.status_code, response.text
def main() -> int:
parser = argparse.ArgumentParser(description="Run basic PolyWeather ops checks.")
parser.add_argument("--base-url", default="http://127.0.0.1:8000")
parser.add_argument("--timeout", type=float, default=8.0)
args = parser.parse_args()
base = args.base_url.rstrip("/")
timeout = args.timeout
report = {"checks": []}
failed = False
try:
_, health = _get_json(f"{base}/healthz", timeout)
ok = str(health.get("status") or "").lower() == "ok"
report["checks"].append({"name": "healthz", "ok": ok, "detail": health})
failed = failed or not ok
except Exception as exc:
report["checks"].append({"name": "healthz", "ok": False, "detail": str(exc)})
failed = True
try:
_, status = _get_json(f"{base}/api/system/status", timeout)
features = status.get("features") or {}
ok = status.get("status") == "ok" and bool((status.get("db") or {}).get("ok"))
report["checks"].append({"name": "system_status", "ok": ok, "detail": {"features": features, "db": status.get("db")}})
failed = failed or not ok
except Exception as exc:
report["checks"].append({"name": "system_status", "ok": False, "detail": str(exc)})
failed = True
try:
_, metrics = _get_text(f"{base}/metrics", timeout)
ok = "polyweather_http_requests_total" in metrics or "polyweather_source_requests_total" in metrics
report["checks"].append({"name": "metrics", "ok": ok, "detail": "metrics exposed"})
failed = failed or not ok
except Exception as exc:
report["checks"].append({"name": "metrics", "ok": False, "detail": str(exc)})
failed = True
print(json.dumps(report, ensure_ascii=False, indent=2))
return 1 if failed else 0
if __name__ == "__main__":
sys.exit(main())