Frequently asked questions

The questions beta testers actually asked, answered without marketing. If yours is not here, the answer is probably in the documentation.

Is the free tier a trial?

No. Five checks at a five-minute interval, indefinitely, with no card. It runs on the same infrastructure and the same two probe regions as every paid plan — there is no separate free-tier engine.

See all plans →

Where is my monitoring data stored?

In France, on Scaleway infrastructure in the fr-par region. A second EU probe region exists for monitoring redundancy and never stores personal data — only check results in transit. A DPA structured on Article 28 GDPR is available to read and download.

Security & data →

How long do you keep my monitoring data?

Check results are kept 90 days and then deleted; output your job attaches to a failed ping is kept 30 days. Both are floors rather than averages — nothing is deleted early, and the security page states how far past each window a record can survive before the mechanism that deletes it next runs. Both windows are the same on every plan, free included: we do not sell retention as an upgrade. Erasure under Art. 17 is immediate and does not wait for either window.

Retention in full →

How do you avoid waking me up for nothing?

A failure has to be confirmed from a second, independent probe region before it pages you, and a check only goes down after N consecutive failures — two by default, configurable per check. A single region losing its network does not produce an alert.

How multi-region confirmation works →

Why did my check go down without sending an alert?

Almost always an unconfirmed alert channel. A channel receives nothing until it is validated: email validates when you click the link sent to that address, webhook, Slack and Discord the moment a test delivery succeeds. This is deliberate — it guarantees alerts never go somewhere nobody confirmed — but it means adding a channel is two steps, not one. Every check shows a delivery log listing what was sent and what was skipped, with the reason.

How alerting works →

What can I be alerted through?

Email, Telegram, Slack and Discord incoming webhooks, and a generic HMAC-signed webhook to any HTTPS endpoint. Channels can be assigned per check or left as the organization default. There is no SMS and no phone call.

Alerting in detail →

Can you monitor a cron job that is supposed to run at 3am on weekdays?

Yes. A cron check takes the crontab line itself, such as 0 3 * * 1-5, plus the timezone your server runs it in. The deadline becomes the next scheduled run plus a grace delay, so a weekday-only job stays quiet over the weekend instead of paging you every Saturday, and a skipped run is noticed inside its own grace window rather than a day later.

Cron and heartbeat monitoring →

How often can a check run?

From every 30 seconds to every 24 hours. The floor depends on the plan: five minutes on Free, one minute on Starter, thirty seconds on Pro and Business.

Plan limits →

Do I get a public status page?

Yes, on Pro and Business. Tower gives you a page at status.okokumo.com with your own slug, where you choose which checks appear as components and name each one for your audience. Nothing else about a check is exposed — not its URL, not its internal name. Business can serve the page from your own hostname with a certificate we obtain and renew.

About Tower →

Can the status page use my own domain?

On Business, yes, for a subdomain such as status.yourcompany.com. You add the hostname, create one CNAME record we show you, and press Verify. An apex domain cannot be a CNAME, so it cannot be pointed at us. Your original okokumo address keeps working either way, so links you have already shared do not break.

About Tower →

Is there an API?

Yes. A versioned REST API under /v1 with per-organization API keys covers creating and managing checks, reading their status and results, and driving status-page incidents. Organization configuration — team, billing, your public address — stays session-only, so a leaked key can narrate an outage but cannot repoint your status page.

Help and documentation →

What does okokumo deliberately not do?

No log aggregation, no APM or tracing, no real-user monitoring, no SMS, and no arbitrary TCP or port checks. The job is to know the moment something you depend on breaks and tell you fast, without crying wolf. Tools that try to be an entire observability platform do that part worse.

Who monitors okokumo?

An independent third-party monitor outside our own infrastructure, which alerts the founder through a channel that does not depend on okokumo being up. Its verdict is published rather than summarised, so you can check it yourself.

Live system status →

How does billing work, and can I leave?

Monthly through Stripe, with French-compliant invoicing: sequential numbering, correct VAT for France, EU B2B reverse charge and non-EU, and the legal mentions French law requires. Upgrades and downgrades are self-serve, and cancelling leaves the account on the free tier rather than deleting it.

Pricing and billing →