The system status page
Before reporting a problem, spend ten seconds here: is the platform itself having an issue? The public status page at https://onflowads.com/status shows the live health of the Onflow Ads backends and any incident notice the team has posted.
No sign-in is required — it is public, and worth bookmarking — and the page keeps working while the rest of the site is in maintenance, since it and its data feed are both exempt, which is exactly when you would want them.
Signed out. The all-green state. The dot on the left and the word on the right always agree; the last-checked time at the bottom is the one thing that moves on its own.
When to check it
- The site feels slow, or actions fail with generic errors.
- The Telegram bot is not responding.
- You have hit a maintenance page and want the bigger picture.
- Before you write in, so your message can say what the status page showed.
Reading the page
One column: a headline card, one row per component, then the freshness line.
The headline
| Headline | Dot | What it means |
|---|---|---|
| All systems operational | Green | Everything that can be judged is passing its check. Sub-line: "Everything is running normally." |
| Some systems are degraded | Amber | At least one judgeable component is failing. Sub-line: "We're aware and working on it." |
| Status unavailable | Red | Your browser could not reach the status service at all. Sub-line: "We couldn't reach the status service. Please retry shortly." |
That state is produced by your own browser when the request fails — it is not the platform reporting itself down. Retry shortly, or try from another network or device.
The components
| Component | What it covers |
|---|---|
| Database | The core storage behind accounts, orders, campaigns, wallets and support conversations. Checked with a trivial query. |
| Cache (Redis) | The fast-access layer behind sessions, rate limits and live availability. Checked with a ping. |
| Telegram bot | Whether the bot's own health endpoint answers. Only checked when the site has been given the bot's health URL. |
Each check has a five-second timeout and they all run at once, so the whole page settles in about the time of the slowest one.
Each row carries one of three labels, shown capitalised:
| Label | Dot | Meaning |
|---|---|---|
| Operational | Green | Healthy right now. |
| Down | Red | Failing its health check at this moment. |
| Not monitored | Grey | This component is not being checked from here. It says nothing about whether it is working. |
Only components that can actually be judged count. A row reading Not monitored is excluded from the verdict, so All systems operational above a Not monitored row is consistent, not a bug.
Incident notices
When the team has an incident message active it appears as an amber banner above the headline card. The same message is also pinned as a dark-red strip across the top of every other page on the site, and it is deliberately injected into the maintenance holding page too, so a locked-out visitor still sees it.
The strip is not dismissible and it stacks above the navigation rather than covering it. An incident notice is written by a person, so it can be present while every component reads Operational — for example when one feature is affected rather than a whole backend.
Freshness
- The page re-checks every 30 seconds on its own, and the answer is never cached.
- The time of the last check is shown at the bottom: "Auto-refreshes every 30s · Last checked …", in your device's local time.
- There is nothing to reload by hand. If you are waiting for a recovery, leave the tab open and watch that line move.
A compact footer carries Platforms, Pricing, About, Contact, Terms, Privacy and Refunds, and ← Back to site sits at the top right.
What the status page cannot see
It reports platform-wide health, so it will show All systems operational even when:
- your sign-in is failing because of your own account state;
- a single order or campaign is stuck;
- a payment has not been credited;
- your browser, an extension, or your network is the problem.
The problem is almost certainly specific to your account or your setup. Work through Troubleshooting & FAQ, then write to the team with the page URL, what you did and the exact error text.
If something goes wrong
| What you see | What it means | What to do |
|---|---|---|
| Some systems are degraded | Our own health checks are failing | Nothing to report — the team sees the same signal. Watch the page; it refreshes itself |
| A component says Down but your feature works | A component can fail a health check while parts of the product carry on | Expect some things to misbehave, not that everything is broken |
| Status unavailable | The status request itself failed | Retry, then try another network or device |
| Telegram bot — Not monitored | The site has not been given the bot's health URL | It says nothing about the bot. Open the bot in Telegram to find out |
| The whole site is a holding page but this page loads | Deliberate — the status page is exempt from maintenance | See Maintenance windows |
| A red strip across every page while status is green | The strip is a human-written incident notice, not an automated readout | Read the message; it says what is affected |
Related
- Maintenance windows — the holding page, and what stays reachable behind it.
- Troubleshooting & FAQ — the problems the status page cannot see.
- Choosing the right support channel — where to go once you know it is not us.