9d6ebbe422
NAS 局域网 IP 因重启被 DHCP 换过两次，磁盘/网络稳定性都不如已经跑着好几个 生产服务的甲骨文机器。整体搬迁：应用 + 数据库都搬走，NAS 只保留 Gitea（这个 仓库的源码托管，未动）。 ## 迁移过程（已核对无损） - MariaDB：NAS 导出（10.11 源库，处理了只有新版本才有的 `/*M!999999` 注释） → 导入甲骨文 MariaDB 10.3.39，**20 张表逐条精确 COUNT(*) 比对完全一致** - 冻结 NAS（停服务）后又 dump 一次核对，确认期间零数据差异，才继续删库 - NAS `garmin_health_lab` 已 DROP DATABASE，备份在本地 `~/Desktop/Work/backups/garmin_health_lab_nas_backup_20260912.sql.gz` - 应用部署到 `/opt/garmin-health-lab`，systemd 单元（`ubuntu` 用户，非 root），和这台机器上的 ai-gateway/auth-hub 同一套约定 - 公网：`https://garmin.zichuan.xyz`，DNS + Caddy 反代 + 自动 TLS，替代原来 `NAS frpc → 甲骨文:8124` 那条隧道（已从 NAS 的 frpc.toml 精确删除对应段， 其它转发未动，改完逐条复检过没打断） - auth-hub 回调地址换成新域名，NAS/旧端口那几条历史回调已清掉 - AI 网关配置改本地回环（网关现在同机了），触发真实生成验证过 ## 一个当场拦下来的风险 甲骨文部署完默认开着自动同步。迁移窗口期两边并行跑时，若两边的调度器同时去 刷新 Garmin 令牌，会撞上按账号计算的 SSO 限流（`GarminHealthLab` 仓库 2026-09-03 那次事故的根因，那次修复花了一整天）。确认账号级 auto_sync 设置 本来是关的、这次算侥幸没撞上——不是设计上的保险，所以迁移期间显式在甲骨文这边 加了 `AUTO_SYNC=false`，直接在运行进程里验证过生效，确认 NAS 已冻结、数据无 缺口后才打开。 ## 文档 / 脚本同步 CLAUDE.md 明确写过"部署位置会变，排障前先查、不要凭记忆"——这次是第二次踩中 同一类问题（上次是"NAS 有没有生产环境"判断错），所以把 CLAUDE.md / PROGRESS.md / README.md / docs/* 里的部署事实全部更新，NAS 时代的 `deploy/` 脚本加废弃 说明保留参考、不删除，新增 `deploy/push_oracle.sh`（当场跑通一次真实部署）。 Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
124 lines
5.8 KiB
Plaintext
124 lines
5.8 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 is co-located with the
|
||
# app on the Oracle box (129.146.26.249, MariaDB 10.3.39), TCP 127.0.0.1:3306,
|
||
# dedicated account (not root). MariaDB treats `user@localhost` and
|
||
# `user@127.0.0.1` as two different accounts — if you create this user by
|
||
# hand, create both host variants with the same password, or TCP connections
|
||
# fail with a password that looks right but is not the one that account has.
|
||
# MARIADB_HOST=127.0.0.1
|
||
# MARIADB_PORT=3306
|
||
# MARIADB_USER=garmin
|
||
# 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 https://garmin.zichuan.xyz/auth/callback — add/remove redirect
|
||
# URIs on that client with /opt/auth-hub's manage_clients CLI, not by editing
|
||
# auth-hub's own database directly.
|
||
AUTH_HUB_REDIRECT_URI=http://127.0.0.1:5500/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.
|
||
# Production (co-located with ai-gateway on the same Oracle box) uses the
|
||
# loopback address — skips Caddy and the public hop entirely, and the token
|
||
# never leaves localhost either way. Use the HTTPS domain instead only when
|
||
# this app runs somewhere else: the token rides in an Authorization header
|
||
# and must not cross the public internet in the clear.
|
||
AI_GATEWAY_BASE_URL=http://127.0.0.1:5100/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
|