682936b0b6
「又被限流了」的根因找到了，不是请求量，是令牌。 `_connect` 里 `refresh_oauth2()` 换来的新 OAuth2 令牌只活在进程内存里—— `save_token` 只在绑定账号时调用过一次。于是每次客户端缓存过期（15 分钟）、 每个 gunicorn worker、每次部署重启，都从库里读回**同一个已过期的令牌**， 然后再做一次真实 SSO 换令牌。而 SSO 端点是按账号限流最狠的那个，社区报告能 封 48 小时（garth #217、python-garminconnect #337）。我今天为了部署重启了 八次服务，每次都清掉缓存。 - `refresh_oauth2()` 成功后 `_persist_token()` 写回。拆出这个函数是因为它和 `save_token` 想要的正好相反：重新绑定要作废现有会话，持久化刷新结果必须 保住刚刚产出它的那个会话 - 写回时不带 garmin_email，否则 upsert 会把绑定邮箱刷成 NULL，数据同步页会 忘记绑的是哪个账号 - 刷新加进程内锁，并在拿到锁后重读一次库：另一个线程刚换过就直接用它的， 不再自己去换一次 - 五条测试盯住这个不变量，包括「冷缓存不该再换一次」（这条如果回归，就是同一 个 bug 再来一遍） 顺带把数据端点也节流了——那是另外一半问题，不是这次的病因，但一天历史要 9 次 调用，730 天全历史 6600 个请求全速打出去，不该指望佳明一直容忍： - services/garmin_throttle.py：代理包住 client，所有调用（含以后新加的）都经 同一个收口，按间隔排队并计数 - 0.5s 是查过的：garmin-data-export 默认 0.15s、garmin-connect-scraper 默认 3s、官方合作方 API 100 次/分钟（0.6s）。依据写在文件顶部 - 单次同步 1200 个请求预算，跑满就干净收尾、下次接着跑（已存的天数本来就跳过） - 运动详情每次最多补 40 条——新账号几百条，不限量就是一次性打光预算 Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
119 lines
5.3 KiB
Plaintext
119 lines
5.3 KiB
Plaintext
# --- Server ---
|
||
# BACKEND_PORT takes precedence over PORT. Prefer it: many tools inject PORT
|
||
# for the frontend, and Flask would otherwise take the React dev server's port.
|
||
BACKEND_PORT=5000
|
||
|
||
# --- Database: sqlite (default) or mariadb ---
|
||
DB_TYPE=sqlite
|
||
# SQLite file (used when DB_TYPE=sqlite)
|
||
DATABASE_PATH=./data/health.db
|
||
|
||
# MariaDB (used when DB_TYPE=mariadb) — production DB on the NAS
|
||
# (192.168.50.64, MariaDB 10.11). Connection is over the socket
|
||
# /run/mysqld/mysqld10.sock (or TCP 127.0.0.1:3306) as root; the socket path
|
||
# only matters when TCP auth is disabled for the app user.
|
||
# MARIADB_SOCKET=/run/mysqld/mysqld10.sock
|
||
# MARIADB_HOST=127.0.0.1
|
||
# MARIADB_PORT=3306
|
||
# MARIADB_USER=root
|
||
# MARIADB_PASSWORD=your_production_mariadb_password
|
||
# MARIADB_DATABASE=garmin_health_lab
|
||
|
||
# --- Auth ---
|
||
# CHANGE THIS in production! Used to sign JWTs (7-day expiry by default).
|
||
JWT_SECRET=dev_secret_change_me
|
||
JWT_EXPIRY_DAYS=7
|
||
|
||
# --- auth-hub OAuth2 provider (centralized SSO) ---
|
||
# See docs/AUTH_HUB_INTEGRATION.md for setup instructions.
|
||
#
|
||
# Base URL of the auth-hub service
|
||
AUTH_HUB_BASE_URL=http://129.146.26.249:5300
|
||
#
|
||
# OAuth2 client credentials (obtain from auth-hub.manage_clients create)
|
||
# NOTE: these must be registered against the auth-hub instance AUTH_HUB_BASE_URL
|
||
# actually points to (dev vs prod are separate databases with separate clients).
|
||
# Put the REAL values in backend/.env (gitignored) — never here.
|
||
AUTH_HUB_CLIENT_ID=your_client_id
|
||
AUTH_HUB_CLIENT_SECRET=your_client_secret
|
||
#
|
||
# Callback URL (must exactly match what's registered in auth-hub). Production
|
||
# registers both the public frp address (129.146.26.249:8124) and the LAN
|
||
# address (192.168.50.64:8124).
|
||
AUTH_HUB_REDIRECT_URI=http://129.146.26.249:8124/auth/callback
|
||
|
||
# --- CORS (comma-separated allowed front-end origins) ---
|
||
# localhost stays in the production list on purpose: CORS is not an auth
|
||
# boundary — every data route requires a valid JWT — so allowing a developer's
|
||
# dev server costs nothing and saves toggling this on every session.
|
||
CORS_ORIGIN=http://localhost:3000,http://localhost:5173
|
||
|
||
# --- AI models (text-only, large context) ---
|
||
# Put REAL keys in backend/.env — that file is gitignored. Never commit keys.
|
||
# Any model whose credentials are absent is skipped automatically.
|
||
|
||
# Self-hosted AI gateway (model id "gateway"). OpenAI-compatible; it fans out
|
||
# over nvidia/gemini/ollama itself and rotates several Gemini keys, so it
|
||
# absorbs single-vendor quota limits. Reached directly, bypassing any local
|
||
# HTTP proxy. NOTE: its NVIDIA upstream is a large reasoning model — replies
|
||
# can take 2-3 minutes, so set AI_TIMEOUT_SECONDS accordingly.
|
||
# HTTPS (Caddy, strips the /ai prefix) rather than http://…:5100 — the token
|
||
# rides in an Authorization header and should not cross the internet in clear.
|
||
AI_GATEWAY_BASE_URL=https://ai.zichuan.xyz/v1
|
||
AI_GATEWAY_TOKEN=
|
||
AI_GATEWAY_MODEL=ai-gateway-auto
|
||
|
||
# Google AI Studio -> "gemini-flash". Free-tier quota is small; 429s are common.
|
||
GEMINI_API_KEY=
|
||
|
||
# NVIDIA NIM -> "llama-70b", "nemotron-49b", "mistral-large".
|
||
# Model ids come from that account's live GET /v1/models — do not guess them.
|
||
NVIDIA_API_KEY=
|
||
# NVIDIA_BASE_URL=https://integrate.api.nvidia.com/v1
|
||
|
||
# Preference order. The first configured model answers; if it fails or times
|
||
# out, the next is tried. Read per request, so changes need no restart.
|
||
AI_MODEL_CHAIN=gateway,gemini-flash,llama-70b
|
||
|
||
# Max days of history sent (CSV-encoded). Trimmed further per model so the
|
||
# payload always fits that model's own context window.
|
||
AI_DAY_BUDGET=365
|
||
|
||
# Measured against the gateway, not guessed: a trivial prompt took 138s end to
|
||
# end, because its primary upstream emits a full chain of thought before the
|
||
# answer. Nothing user-facing blocks on this (the briefing generates in a
|
||
# background thread), but the timeout still has to clear the real latency.
|
||
AI_TIMEOUT_SECONDS=300
|
||
# Output cap. Reasoning models spend part of it thinking before they answer;
|
||
# entries that need more declare their own budget in services/ai.py.
|
||
AI_MAX_TOKENS=1024
|
||
|
||
# Output cap for the AI coach (晨报 / 趋势归因 / Copilot). Larger than
|
||
# AI_MAX_TOKENS above: the same reasoning trace is spent from this budget
|
||
# before the answer starts, and at 1024 the reply was all thinking with the
|
||
# JSON truncated away.
|
||
AI_COACH_MAX_TOKENS=4000
|
||
|
||
# --- AI coach job queue ---
|
||
# Three projects share the gateway, so going too high causes 502s.
|
||
AI_JOB_CONCURRENCY=2
|
||
AI_JOB_GAP_SECONDS=5
|
||
# Set AI_JOBS=false to stop consuming entirely (screens then show the computed
|
||
# figures with no model reading).
|
||
AI_JOBS=true
|
||
|
||
# --- Garmin 请求节流 ---
|
||
# 一天的历史要 9 次 API 调用（_extract_daily 6 + daily_extras 3），短同步再加
|
||
# 5 次曲线，每条没有详情的运动 1 次。730 天全历史 ≈ 6600 个请求。以前是能发多
|
||
# 快发多快。
|
||
#
|
||
# 0.5 秒的依据（2026-09-03 调研，见 services/garmin_throttle.py 顶部注释）：
|
||
# sirredbeard/garmin-data-export 默认 0.15s、evg656e/garmin-connect-scraper
|
||
# 默认 3s、佳明官方合作方 API 是 100 次/分钟（合 0.6s）。
|
||
GARMIN_MIN_INTERVAL_SECONDS=0.5
|
||
# 单次同步的请求预算。跑满就干净收尾，下次接着跑（已存的天数会跳过）。
|
||
# 1200 × 0.5s ≈ 10 分钟，够覆盖四个月历史。
|
||
GARMIN_REQUEST_BUDGET=1200
|
||
# 每次同步补多少条运动详情。新账号有几百条，不限量就是一次性打光预算。
|
||
GARMIN_DETAILS_PER_SYNC=40
|