Uptime SLA

SupaHost's monthly uptime commitment per plan (Pro 99.5%, Business and Enterprise 99.9%), how downtime is measured, the service credits, what's excluded, and how to claim.

SupaHost commits to a monthly uptime for each instance on a paid plan, and gives you a credit on your bill if it misses. This page summarises the commitment. The full terms are in the Service Level Agreement, which is part of the Terms of Service and takes precedence over this summary.

The commitment

Plan Monthly uptime Allowed downtime in a 30-day month
Builder No commitment
Pro 99.5% about 3 hours 36 minutes
Business 99.9% about 43 minutes
Enterprise 99.9%, or as set in your order form about 43 minutes

The commitment applies to each instance separately, for each calendar month (UTC) the instance is on a paid Pro, Business or Enterprise plan. It doesn't apply during a free trial.

How uptime is measured

SupaHost's monitoring checks every instance once a minute from outside it. A check fails when the instance doesn't answer over HTTPS, its services report that they aren't ready, or its management agent stops reporting.

Downtime starts at the first of three or more failed checks in a row and ends at the next successful check. One or two failed checks on their own don't count.

These are the same checks behind the instance down and recovered emails. Keep Downtime alerts on (the default) in Dashboard → Settings → Email notifications so you know when an incident starts and ends; see Monitoring.

Service credits

If an instance misses its commitment in a month, you can claim a credit on that instance's plan fee for the month:

Monthly uptime Credit
Below the commitment, but at least 99.0% 10%
Below 99.0%, but at least 95.0% 25%
Below 95.0% 50%

Credits are applied to your next invoice and aren't paid out in cash. They're worked out on the plan fee only, not on add-ons or usage, and capped at 50% of the plan fee for the month.

What doesn't count as downtime

  • Scheduled maintenance inside your instance's maintenance window, or announced at least 48 hours ahead, up to 4 hours a month.
  • Your own actions, or ones you ask SupaHost to take: restores, plan changes, restarts, upgrades you start, deleting or changing projects, and changes over SSH, MCP or SQL.
  • Running out of resources because of your workload (CPU, memory, disk or connections), including reaching a plan quota.
  • Third-party services you configure, such as your own SMTP or SMS provider, OAuth providers, DNS for your custom domains, and your Edge Function code.
  • Suspension under the Terms of Service, for example for non-payment.
  • Events beyond reasonable control, such as wide-scale internet or cloud region failures.
  • Beta or preview features, as marked in the dashboard or docs.

Claiming a credit

Email [email protected] within 30 days of the end of the month. Include the instance name and the dates and times of the downtime. The down and recovered emails, and the incidents in Activity, record those times for you.

What the SLA doesn't change

The SLA is a commitment with credits, not a different architecture. Each instance is still one VM:

  • There's no multi-server failover and no read replicas. If the VM or its zone has a problem, your projects are down until it recovers.
  • Plan for brief outages in your app: retry failed requests, and show a clear message instead of a blank screen.
  • SupaHost is in public beta, and doesn't hold SOC 2, ISO 27001 or HIPAA certifications.

Updated 11 November 2026 · View as Markdown · Something wrong or missing? Tell us