Operations
Operational status.
Measured, not asserted.
There is no traffic light on this page, and that is on purpose. This page is generated at build time and served unchanged afterwards; a green dot on it would show the state at publication and would still be green during an outage. So the measurement lives where it belongs: on a page of its own that re-checks every thirty seconds in your browser.
How it is measured
The status page calls each service individually and reads the answer. That happens in the visitor's browser, not on one of our servers: there is no cache in between whose contents could go stale, and no place where a result could be quietly improved.
Two of the services answer with structured data rather than a bare reachability signal. The API gateway reports in the same response whether the database, the cache and the model routing are working; the collaboration server reports its service identifier and version. The rest are checked for reachability.
- A check every thirty seconds for as long as the status page stays open
- Eight services, each named and each rated on its own
- The API health endpoint answers as JSON and can be polled by your own monitoring system
Eight services under observation
The names match the ones on the status page, so that your observation and your report to us use the same words.
| Address | What is checked | |
|---|---|---|
| API gateway | api.simosphereai.com | Reachability plus database, cache and model routing from the same response |
| Admin dashboard | app.simosphereai.com | Reachability |
| Operator console | console.simosphereai.com | Reachability |
| Signup portal | onboarding.simosphereai.com | Reachability |
| Web workspace | workspace.simosphereai.com | Reachability |
| Collaboration server | collab.simosphereai.com | Reachability plus the reported service identifier and version |
| File storage | nc.simo-online.com | Reachability behind the access gate |
| The status page itself | status.simosphereai.com | Reachability, so you can tell whether the measurement itself is still running |
When something does not work
Four steps, in this order. The first two take a minute and may save you the report altogether.
Open the live status
On status.simosphereai.com you can see in one pass whether one service is affected or several. That already narrows things down: if only the workspace is down, the API usually keeps serving.
Query the health endpoint
api.simosphereai.com/health answers as JSON with a status field and a timestamp. Copy the response into your report. Then the time is on record instead of estimated.
Report the outage
Write to [email protected]. Four details help: the service affected, the time including time zone, what you did, and what happened instead.
Report security incidents separately
If you suspect a vulnerability, unauthorized access or a data leak, send it to [email protected] rather than to the general inbox. That address is handled on a separate path.
Common questions about operations
Why is there no traffic light on this page?
Because it would offer false comfort. This page is static: its content is created at build time and does not change afterwards. An indicator on it would show the state from back then. Green, precisely during an outage. The measurement belongs where it happens at the moment you look.
How current is the live status?
The check repeats every thirty seconds for as long as the page stays open and queries the services directly. What you see is therefore at most thirty seconds old.
Can I pull the state into my own monitoring system?
Yes. api.simosphereai.com/health answers as JSON with a status field and a timestamp and can be polled like any other health endpoint: no login, no key.
Does the status cover this website as well?
No. This site runs separately from the platform. If you are reading this sentence, the site is being served; that says nothing about the state of the platform services.
Where do I find availability commitments?
Not on this page, but in § 11 of the terms and conditions: the platform is committed to 99.5 percent availability on an annual average, and planned maintenance windows are announced at least 48 hours in advance and do not count as downtime. Anything beyond that, such as response times, is set out in your own contract. This page describes the measurement, not the commitment.
Something not reachable?
Then tell us once too often rather than once too rarely. A report with a time and a service name takes five minutes to write and can save half a day.