Last updated: 2026-05-26
Service Level Agreement
This SLA covers the Sentinel public API at api.pauseapi.app and the dashboard at app.pauseapi.app. Effective for any customer on the Pro plan and above; Free tier is offered on a best-effort basis without SLA commitments.
Uptime commitment
| Plan | Monthly Uptime Target | Measurement window |
|---|---|---|
| Free | Best effort (no SLA) | — |
| Pro | 99.5 % | Calendar month, UTC |
| Enterprise | 99.9 % (negotiable to 99.99 %) | Calendar month, UTC |
What counts as downtime
Sentinel is considered down for a given minute when both of the following hold:
- At least three consecutive synthetic probes against
https://api.pauseapi.app/healthfrom at least two regions return non-2xx or fail to connect. - The failure is not caused by (a) maintenance announced ≥ 72 hours in advance, (b) customer-side networking, (c) abuse / DOS attempts, or (d) a Force Majeure event.
Uptime % for a month is calculated as (total_minutes − down_minutes) / total_minutes × 100, rounded down to two decimals.
Service credits (Pro)
If we miss the monthly uptime target, you can claim a credit on your next invoice equal to the percentage shown:
| Monthly Uptime achieved | Credit on next invoice |
|---|---|
| ≥ 99.5 % | 0 % |
| 99.0 % – 99.49 % | 10 % |
| 95.0 % – 98.99 % | 25 % |
| < 95.0 % | 50 % |
Credits are not refunds. The maximum monthly credit is 50 % of that month's invoice. Credits expire 12 months from issue.
How to claim a credit
Email support@regengine.co within 30 days of the calendar month in question. Include:
- Your tenant ID (
ten_…) - The month being claimed
- Optional: your own monitoring data, if relevant
We review and respond within 5 business days. We publish historical uptime on the status page — if our number says we hit the target, we'll explain. If our number disagrees with your data, we'll investigate together.
Maintenance
Planned maintenance windows are announced at least 72 hours in advance via:
- Email to the account-owner address on file
- Notice on the status page
We aim to schedule maintenance during low-traffic windows (Saturdays 06:00 UTC).
Out of scope
- Webhook delivery success to your endpoint — we retry 3 times with exponential backoff (1 s, 4 s, 16 s) but cannot guarantee delivery if your endpoint is unreachable.
- Notification delivery via Resend (email) or Twilio (SMS) — their SLAs govern. We pass through any vendor-side outage.
- Free-tier service is offered without uptime guarantees.
Force Majeure
Outages caused by events outside our reasonable control (regional cloud-provider failures, cable cuts, sanctioned-network actor attacks, governmental action) are excluded from uptime calculations for the duration of the underlying event plus 4 hours of recovery.
Changes to this SLA
We may revise this SLA with at least 30 days' notice via the same channels used for maintenance announcements. Existing annual contracts are honored on the SLA in effect at signature for the duration of the contract term.
Questions? Email support@regengine.co.