Uptime over the past 90 days. View historical uptime.
The REST API behind every TimerOS client. Sign-in, time entries, everything.
The TimerOS dashboard in a browser, at app.timeros.ai.
The branded space your clients sign in to, on your workspace subdomain.
timeros.ai — marketing pages, registration, and plan management.
The update feed and installer downloads the Windows app depends on.
Past incidents
Full historyNo incidents reported in the last 15 days.
Subscribe to updates
Atom / RSS feed
Every incident and update, in any feed reader.
https://status.timeros.ai/history.atom
JSON API
Machine-readable current status. CORS is open, so you can call it from a browser.
https://status.timeros.ai/api/summary.json
Not offered yet. Adding a subscriber list means processing personal data, so it ships with the privacy notice updated, not before. Until then, the feed above carries the same information — or write to support@timeros.ai.
Infrastructure we depend on
TimerOS runs on these providers. Their outages can take parts of TimerOS with them, and the SLA excludes them from our availability commitment — so when something here is red, one of these is worth checking.
How this is measured
This page is an independent monitor. It runs on Cloudflare and shares no infrastructure with the services it checks — no shared host, no shared database, no dependency on the TimerOS API — so an outage cannot take the page reporting it offline.
- Interval
- Every 60 seconds, from Cloudflare's network.
- Failure
- A check fails on a non-2xx response, a timeout, or a response that does not contain what it should. 2 consecutive failures are needed before a component is shown as disrupted, so a single dropped request does not turn the page red.
- Uptime
- The percentage shown above counts every failed check, however brief — the conservative reading, which can only make us look worse than the contract requires. Measured time only: if the monitor itself did not run, that period is excluded from the calculation and shown as "no data" rather than assumed healthy. Scheduled maintenance is excluded on both sides.
- SLA downtime
- A separate, narrower figure: only continuous failures lasting 5 minutes or longer count as Unavailability under §1 of the SLA, and only that figure is measured against the 99.5% non-binding monthly target. It is published per month on the history page. So a red bar can sit next to a high uptime percentage: something failed, but not for long enough to count under the contract.
- Caching
- The website, web app and portal sit behind Cloudflare's cache, so a check may be answered from the edge. That is intentional: a visitor would be served those same bytes. The API is not proxied and is always measured against the origin.
- Vantage point
- A single one. This page can tell you TimerOS is unreachable from Cloudflare's network; it cannot tell you whether your own network is the problem.
- Scope
- Components marked Informational are monitored but are not covered by the SLA's availability commitment.
Continuous monitoring began . Days before that are shown as "no data" and are not counted in any uptime figure.