Scope
This Service Level Agreement (“SLA”) applies to every dedicated server we provide, from the moment it is delivered, and is part of our Terms of Service. Commitments are measured per server and per calendar month.
Network availability: 99.9%
We commit to 99.9% network availability per calendar month. The network is unavailable when your server’s port cannot exchange traffic with the internet for five consecutive minutes or more because of a fault in our network, a facility outage or a hardware failure on our side.
Availability is measured by our own monitoring, which probes every server’s uplink every minute from outside the facility, and by the incidents you report to support.
Hardware replacement
Spare parts are kept on site in every location and technicians work around the clock.
- Failed disk: replaced within 2 hours.
- Any other component (memory, power supply, motherboard, NIC, CPU): replaced within 4 hours.
The clock starts when the fault is confirmed — by our monitoring, by our diagnostics, or by the ticket in which you report it with the evidence we ask for (a SMART report, a kernel log). For hot-swap disks we can replace a drive without powering the server down; for other parts we need your go-ahead for a short intervention.
Support response
Support is staffed by engineers 24 hours a day, every day of the year. Our median first reply is under 15 minutes. For a server that is down or unreachable, we aim to reply within 1 hour at the latest.
Service credits
If we miss a commitment, you receive a credit calculated on the monthly price of the affected server:
| Monthly network availability | Downtime in a 30-day month | Credit |
|---|---|---|
| 99.9% or more | up to 43 minutes | — |
| 99.5% to 99.9% | 43 minutes to 3 h 36 min | 5% |
| 99.0% to 99.5% | 3 h 36 min to 7 h 12 min | 10% |
| 98.0% to 99.0% | 7 h 12 min to 14 h 24 min | 25% |
| Below 98.0% | more than 14 h 24 min | 50% |
- Late hardware replacement: 5% of the monthly price for every started hour beyond the target, up to 50%.
- Credits for one server in one month are capped at 100% of its monthly price.
- Credits are added to your account balance and used for your next invoices. They are not paid out in cryptocurrency.
What is not covered
- Planned maintenance announced at least 48 hours in advance, limited to four hours per month.
- Problems caused by your software, configuration or actions, including reboots, firewall rules and reinstalls.
- Suspensions under the Acceptable Use Policy or for non-payment.
- Attacks so large that they threaten the network as a whole, when traffic to the targeted address has to be dropped to protect other customers. We tell you immediately when this happens.
- Failures outside our network, such as routing problems at a remote provider, and events beyond reasonable control (natural disasters, war, a regional power grid failure).
How to claim a credit
- Open a ticket within 14 days after the end of the month concerned.
- Include the server, the time window and, if you have it, what you observed (monitoring graphs, traceroutes).
- We check the claim against our monitoring and reply within 5 business days. Accepted credits are added to your balance straight away.
We also credit you proactively when our own monitoring shows an outage that breaches this SLA.