Service level agreement
Last updated: 22 July 2026
1. What this covers
This agreement sets out the availability we target for Nintech hosting, how we support you, and what happens when we fall short. It applies to the managed hosting and infrastructure services in your account and is read together with our Terms of Service; bespoke engagements are governed by their own written proposals. We are a small, senior-run operation, so everything below is something one accountable engineer can actually deliver — there are no promises here we cannot keep.
2. Uptime targets
We target the monthly availability published on the Hosting page for each tier: 99.9% for Cloud VPS, game servers, secure email, and VPN; 99.95% for GPU compute and Managed Scale; and 99.99% for Managed Enterprise. Where a service is not named in one of those tiers, we aim for the same 99.9% order of availability, but the numeric targets we formally commit to are the three published tiers. Availability is measured over a calendar month, for the service itself reachable from our network. As the Terms put it, these are commitments of effort and priority, not an insurance policy.
3. What counts as downtime
Downtime is a period in which your service is wholly unavailable because of a fault in the infrastructure we operate — the server unreachable, the platform not responding — confirmed either by our monitoring or by a reproducible check. It is counted from when we detect it or when you report it in a ticket, whichever comes first, until service is restored. Announced maintenance and the exclusions listed below do not count towards it.
4. How we monitor it
Availability is watched by automated monitoring, not by marketing copy. The platform runs scheduled health checks against every server through the provider's API, and heartbeat monitors over the services that carry them — SSL renewal and mail delivery among them — recording the outcome on every run; an operations dashboard flags anything unhealthy or a monitor that has gone quiet. This is automated detection with a senior engineer responding, not a staffed round-the-clock operations centre, and we will not pretend otherwise. The optional Managed monitoring add-on adds uptime, latency, and resource alerts watched by our team, and our Managed SRE engagement adds a 24/7 on-call rotation for clients who need one.
5. Maintenance windows
Planned maintenance — patching, upgrades, and migrations — is announced in advance through your ticket or by email, scheduled for quieter hours where we can, and does not count against the uptime target. Emergency maintenance to close a security hole or stop an active fault may happen with little or no notice; we keep it as short as we can and tell you afterwards what happened and why.
6. Support response
Support is ticket-based: open one from your portal, or by email if you contacted us as a guest. We target a first response from a real engineer within the business-hours bands published on the Hosting page — next business day for VPS, game, email, and VPN; four business hours for GPU and Managed Scale; and one hour for Managed Enterprise, with escalation to a named engineer. A first response means a human engineer engaging with your ticket, not an automated acknowledgement. Business hours are UK working hours, Monday to Friday, excluding public holidays. The Priority support add-on brings same-day responses and escalation to a senior engineer on any plan.
7. Service credits
If a service misses its monthly uptime target because of a fault on our side, you can claim a service credit. Open a ticket within 30 days of the month in question; we verify the downtime against our monitoring, and where it is confirmed we apply a pro-rata credit for the affected service — the share of the monthly fee for the time it was unavailable beyond its target — against a future invoice. Credits are applied to your account rather than paid in cash, and are the sole remedy under this agreement for missed availability. We keep this deliberately simple: a credit we can calculate and honour beats a formula we cannot.
8. Exclusions
The uptime target and service credits do not apply to downtime caused by anything outside our own infrastructure — an outage at DigitalOcean, a registrar, a payment provider, or the wider internet; by your own configuration, code, or software, or a workload that exhausts the resources of its plan; by suspension or termination for non-payment or a breach of the Acceptable Use Policy; by scheduled or emergency maintenance as described above; by denial-of-service attacks beyond the mitigations included with your plan; or by events beyond our reasonable control. Beta and preview features are provided as they are, without an availability target.
9. Changes and governing law
We may update this agreement as our services change; material changes carry the date below, and the targets on the Hosting page remain the reference. It is governed by the law of England and Wales and read together with our Terms of Service.