OpenUptime

Changelog

What changed in OpenUptime, newest first. Everything in the Community edition also ships in Cloud; Cloud adds the features that only make sense in a hosted service.

更新日志

OpenUptime 的变更记录,按时间倒序。社区版的所有改动同样会进入 Cloud 版;Cloud 版另有只适用于托管服务的功能。

Communitymain · MIT

The open-source edition you run yourself, with no plans and no limits. Every change lands here first.你自己部署的开源版本,没有套餐,也没有额度限制。所有改动都先在这里落地。

Cloudcloud

The hosted service. It includes everything in Community, plus plans, billing, hosted probes and the features marked Cloud only.托管服务。包含社区版的全部功能,另有套餐、计费、托管探针,以及标注为「仅 Cloud」的功能。

OpenUptime can be installed from the browser, and an open tab tells you when a new version is ready.可以从浏览器把 OpenUptime 安装成应用;有新版本时,已打开的页面会提示你刷新。

  • Use “Install” in your browser to get OpenUptime in its own window, with an icon on your desktop or home screen.在浏览器中选择「安装」,即可在独立窗口中使用 OpenUptime,并在桌面或主屏幕上留下图标。
  • An open tab checks for a new version every 6 hours and offers a Refresh button when there is one.已打开的页面每 6 小时检查一次新版本,有更新时会提示并提供「刷新」按钮。
  • The app's own files are downloaded in the background when a new version is found, so refreshing is quick and the app shell still opens without a network.发现新版本时,应用自身的文件会在后台提前下载,刷新后很快就能打开;没有网络时应用外壳也能打开。
  • Nothing from the API is stored on your device: monitor data and sessions always come from the server.API 数据不会缓存在设备上:监控数据和登录状态始终来自服务器。

A monitor can be checked from more than one place, so a bad route does not look like your site being down.一个监控项可以从多个地区发起检测,某条线路出问题时不会被误判为站点宕机。

  • Pick regions per HTTP or TCP monitor. A run fails when most of the regions that answered failed; a probe that cannot be reached is left out of the vote.为 HTTP 或 TCP 监控项选择检测地区。多数已应答地区失败才判定本次失败;无法连通的探针不参与投票。
  • Probes run as a Docker image, a plain Node server or on Vercel, and there is a new Multi-region checks docs page.探针可作为 Docker 镜像、Node 服务或在 Vercel 上运行,并新增了「多区域检测」文档页。
  • The monitor page shows one multi-region check as one run, with uptime and latency for each region.监控详情页把一次多区域检测显示为一次运行,并给出各地区的可用率与延迟。
  • With no plan there is no limit on regions per monitor.没有套餐限制时,每个监控项可使用的地区数量不设上限。
On CloudCloud 版Five hosted regions: us-east, us-west, eu-central, ap-southeast and ap-northeast. Free checks from one region, Lite and Starter from up to three, Team from up to five. Each region counts as one check.托管 5 个地区:us-east、us-west、eu-central、ap-southeast、ap-northeast。Free 可选 1 个,Lite 与 Starter 最多 3 个,Team 最多 5 个。每个地区计 1 次检测。
Read the multi-region docs阅读多区域检测文档 →

A pass over the API ahead of the open-source release.开源发布前对 API 做的一轮安全加固。

  • Private and internal check targets are refused by default.默认拒绝检测私有网络与内部地址。
  • Channel credentials and monitor headers are hidden from people who cannot manage them.通知渠道的凭据与监控项请求头,对无管理权限的成员隐藏。
  • Sign-in and sign-up endpoints are rate limited, and email sign-up can use Cloudflare Turnstile.登录与注册接口增加频率限制,邮箱注册可启用 Cloudflare Turnstile。
  • Destructive and configuration changes need an owner or admin.删除与配置类操作需要所有者或管理员权限。
  • At most one automatic incident is open per monitor and kind, and each incident is resolved once.每个监控项的每种事件类型最多同时存在一个自动事件,且每个事件只会被恢复一次。

OpenUptime is MIT licensed and runs without limits on your own infrastructure.OpenUptime 以 MIT 协议开源,可在你自己的基础设施上无限制运行。

  • One Edition interface decides what a workspace may use; the community edition is unmetered.通过统一的 Edition 接口决定工作区可用的额度;社区版不计量、不设限。
  • README, a docs index, and self-host guides for Cloudflare and Docker.新增 README、文档索引,以及 Cloudflare 与 Docker 的自托管指南。
Self-host guide自托管指南 →

Fixes and polish around the app.应用内的若干修复与打磨。

  • Switch workspace or project without reloading the page.切换工作区或项目时不再刷新页面。
  • MCP sign-in no longer fails when a token is refreshed.修复 MCP 授权在刷新令牌时失败的问题。

Team workspaces can check a monitor every 30 seconds, for services where a minute of downtime is already too long.Team 工作区可以把监控项的检测间隔缩短到 30 秒,适合一分钟宕机都无法接受的服务。

  • Pick 30 seconds in the interval list of a monitor; every other plan starts at 60 seconds (Free at 5 minutes).在监控项的间隔下拉里选择 30 秒;其他套餐最短 60 秒(Free 为 5 分钟)。
  • A 30-second monitor uses twice the check credits of a 60-second one, and the usage page counts it that way.30 秒监控项消耗的检测额度是 60 秒的两倍,用量页面也按此计算。
See the plans查看套餐 →

Get a notice before an HTTPS certificate expires, instead of finding out from a browser warning.HTTPS 证书到期之前先收到提醒,而不是等访客看到浏览器警告才知道。

  • HTTPS monitors on port 443 are checked for certificate expiry automatically; an expiring certificate sends a notice, not an outage incident.443 端口的 HTTPS 监控项会自动检查证书到期时间;证书即将到期时发送提醒,而不是宕机事件。
  • A monitor of type SSL watches the certificate of a hostname on its own and opens an incident when the certificate has a problem.「SSL」类型的监控项可以单独监控某个域名的证书,证书出现问题时会开启事件。
  • An SSL monitor uses one check credit per check. Available on Lite and above.SSL 类型的监控项每次检测消耗一个检测额度。Lite 及以上套餐可用。

Serve a status page from your own hostname, such as status.example.com, instead of the shared address.用你自己的域名(例如 status.example.com)访问状态页,而不是共享地址。

  • Add the hostname on the status page, create the DNS record shown there, and the page goes live once the domain is verified.在状态页上添加域名,按提示创建 DNS 记录,域名验证通过后页面即可访问。
  • Available on Team. When a workspace leaves Team, its custom domains are released and the status pages stay reachable at the shared address.Team 套餐可用。工作区离开 Team 后,自定义域名会被释放,状态页仍可通过共享地址访问。
See the plans查看套餐 →

The hosted service now has Free, Lite, Starter and Team plans, with a usage page that shows what is left in the period.托管服务现在提供 Free、Lite、Starter 与 Team 四档套餐,并有用量页面显示当前周期的剩余额度。

  • Each plan sets active monitors, check credits, the shortest interval, regions per monitor, and how long results and alert logs are kept.每档套餐决定可用的监控项数量、检测额度、最短间隔、单个监控项的区域数,以及检测结果与告警记录的保留时长。
  • Subscribe, change plan and manage payment details from Plan & usage; payments are handled by Stripe.在「套餐与用量」页面订阅、变更套餐并管理支付信息;支付由 Stripe 处理。
  • Accounts must verify their email before they can create monitors or status pages.账号需先验证邮箱,才能创建监控项和状态页。
See the plans查看套餐 →

Last updated 2026-10-12最后更新:2026-10-12