Service Level Agreement

This SLA applies to all services we sell: virtual servers and Cloud Apps subscriptions. It covers the network, the hardware your service runs on, and the availability of the control panel.

It does not cover the software running inside your service. A server whose operating system will not boot because of a change you made, or an application that is failing for its own reasons, is not downtime under this agreement — though we will usually help you work out which it is.

The physical network will be available 99.9% of the time in any calendar month, measured from outside our racks. We further commit that our own network hardware will not be responsible for more than 0.3% average packet loss, or more than 0.5 ms of average packet delay variation.

The machines running your service will be available 99.9% of the time in any calendar month. Time during which a service cannot run because of a hardware fault on its host is treated identically to network downtime for the purposes of this agreement.

• We aim to respond to a new ticket within 2 hours during business hours, defined as 09:00–21:00 US Eastern, Monday to Friday.
• A ticket opened outside those hours is treated as opened at the start of the next business day for the purposes of this target.
• Where a delay beyond that target is our fault, the affected time is credited on the same basis as downtime.
• A performance problem we have identified as ours will be resolved within 24 hours of that determination.
• Suspected outages and security issues are handled outside business hours as fast as we are able. We are a small team and we would rather tell you that than promise a night shift we do not staff.

• Scheduled maintenance. Each maintenance event is announced in advance and describes how services will be affected.
• Denial-of-service attacks aimed at your service. This SLA is suspended for the target of an attack for its duration; it continues to apply to every other customer and service during that attack.
• Anything outside our direct control, including upstream network failures beyond our provider's edge, and acts of God.
• Downtime caused by your own configuration, software, or resource exhaustion inside your service.
• Your inability to reach the service because of a fault on your own connection.
• Suspension of a service under the Terms of Service, including for nonpayment or acceptable-use violations.

Credits are proportional. Take the length of credited downtime, divide it by the length of the service period, and multiply by the cost of the service for that period.

Example. A service costing $20.00 per month that suffers one hour of credited downtime receives (1 hour ÷ 720 hours) × $20.00 = $0.03 in account credit.

• Credits are issued as account credit, not as a refund to a payment method.
• A credit is capped at the monthly cost of the affected service. This agreement does not pay out more than you paid.
• If the total credit for a service in a month would be less than $1.00, no credit is issued for that service.
• Credit is the sole and exclusive remedy for a failure to meet the commitments in this agreement.

The owner of the affected service must request the credit through a ticket in the customer portal within 30 days of the incident, describing the service affected and the period concerned. We will check it against our own monitoring and apply any credit due to your account balance.