daniel.w
/
thread-master
1
0
Fork 0
Code Issues Pull Requests 1 Packages Projects Releases Wiki Activity
35 Commits 6 Branches 0 Tags 33 MiB
Go 47.3%
TypeScript 42.3%
CSS 6.5%
Shell 2.3%
JavaScript 1.2%
Other 0.4%
Go to file
王性驊 8c1be42f4a prod: throttle persona scrape jobs, cap worker CPU, add job-failure alerting + offsite backup 
- job: cap concurrent persona_analyze_account (real headless-Chromium scrape)
  jobs per owner to 1 in flight via CountActiveByOwnerAndTemplate; rejects
  with the new ErrTooManyActive (429) instead of letting a member queue up
  many scrapes and pin the shared 1-vCPU host, or double-schedule the same
  persona while a previous run is still executing.
- harbor-worker@.service: CPUQuota=70% so nginx/gateway keep headroom while
  a scrape job briefly saturates Chromium.
- backup.sh: default BACKUP_DIR now matches the systemd unit (/var/backups/harbor)
  instead of the stale /var/backups/haixun default.
- add harbor-job-health.timer (every 5m): queries Mongo for jobs that failed
  in the last 15 minutes and writes a node-exporter textfile-collector
  metric (haixun_job_failures_recent), with new Prometheus alerts
  (PersonaScrapeJobsFailing / BackgroundJobsFailing / JobHealthMetricStale)
  so a stuck worker surfaces on its own instead of only via member reports —
  this would have caught the recent Playwright headless-shell bug directly.
- add harbor-offsite-backup.timer + backup/offsite-sync.sh: pushes
  backup.sh's local backups to an off-host restic repository (client-side
  encrypted). Installed and enabled by default but a safe no-op until
  RESTIC_REPOSITORY/RESTIC_PASSWORD are set in harbor.env.

Co-authored-by: Cursor <cursoragent@cursor.com>
 		2026-07-17 17:38:36 +00:00
.grok/skills/spec-driven fix all 2026-07-09 03:45:28 +00:00
apps prod: throttle persona scrape jobs, cap worker CPU, add job-failure alerting + offsite backup 2026-07-17 17:38:36 +00:00
deploy prod: throttle persona scrape jobs, cap worker CPU, add job-failure alerting + offsite backup 2026-07-17 17:38:36 +00:00
docs/product fix frontend 2026-07-13 03:18:08 +00:00
.gitignore fix frontend 2026-07-13 08:59:13 +00:00
AGENTS.md fix frontend 2026-07-10 12:54:45 +00:00
README.md fix frontend 2026-07-10 12:54:45 +00:00

README.md

Thread-Master / 巡樓（重構主線）

自 2026-07-09 起：舊程式已整包封存至 old/，從零重寫。

現在讀哪裡

路徑 說明
docs/product/haixun-console/ 產品需求 → spec → plan → tasks（真相來源）
old/ 舊 monorepo 只讀封存，不在上面開新功能
.grok/skills/spec-driven/ /spec-driven 流程 skill

目標結構（建立中）

thread-master/
  apps/web/          # Harbor Desk 前端（Phase B，mock 優先）
  services/api/      # 新後端（Phase A/C；名稱以 plan 為準）
  docs/product/      # 產品文件
  old/               # 舊碼封存

交付順序

A 地基（登入 / 權限 / Threads / Job / 通知）
  → B 新前端 + Mock（流程與風格先過關）
    → C 接真後端
      → D 業務加深（海巡等）

前端（apps/web）

cd apps/web
npm install
npm run dev      # 本機開發，監聽 0.0.0.0:5173（區網可連）
npm run build    # 型別檢查 + 生產建置

部署（dev 一鍵清乾淨重裝）

見 deploy/README.md。

cp deploy/.env.example deploy/.env.dev   # 首次：填密鑰
./deploy/scripts/redeploy.sh --yes --kill-orphan
# 可選 nginx：./deploy/scripts/install-nginx.sh

產品任務進度：docs/product/haixun-console/tasks/INDEX.md（目前從 T001 起）。

開發流程

/spec-driven haixun-console
# 或依 tasks 逐題實作，例如「做 T002」

查舊行為請到 old/backend/（只讀）。