Service Level Agreement

This Service Level Agreement (SLA) sets out our network availability commitment and the credits you can claim if we miss it. It applies to active, paid services and forms part of our Terms of Service. Credits under this SLA are your exclusive remedy for downtime.

1. The 99.9% commitment

We target 99.9% monthly network availability for each service. "Network availability" means your service can reach and be reached over our network. Over a 30-day month, 99.9% allows for about 43 minutes of qualifying downtime. Availability is calculated per service and per calendar month. This commitment covers our network reaching your service; it does not measure your own operating system or applications running on the server, which are your responsibility.

2. How downtime is measured

Downtime is a period during which the service is unreachable over our network due to a fault within our control. We measure it from two sources:

• Our monitoring: the internal systems that watch our network and hosts; and
• Customer reports: the timestamps you give in your claim.

Downtime starts when a qualifying outage is first detected or credibly reported and ends when the service is restored. We reconcile the two sources and use them together to determine the affected period. Brief packet loss or latency that does not make the service unreachable is not counted as downtime.

3. Service credits

If monthly network availability for a service falls below 99.9%, you may claim a credit against that service's monthly fee, on this scale:

Credits are added to your account balance and can pay for any server or renewal. Credits are not paid out in cash or crypto.

4. Maximum credit

The maximum credit for any service in any single calendar month is 50% of that service's monthly fee, regardless of the number or length of incidents in that month.

5. How to claim

To claim a credit, open the service's page in the client area and choose Claim an SLA credit under Server actions, within 7 days of the incident. The claim is attached to the service, so its ID goes with it. Include:

• the date, start and end times, and time zone of the downtime; and
• a short description of what you observed.

We compare your report against our monitoring and confirm the eligible credit. Claims made after the 7-day window are not eligible.

6. Exclusions

The following do not count as downtime and are not eligible for credits:

• Scheduled maintenance announced at least 48 hours in advance.
• DDoS attacks beyond our mitigation capacity.
• Customer-caused issues: your software, configuration, operating system, or applications, and unreachability caused by your own firewall or credentials.
• Suspensions for violations of the Acceptable Use Policy or non-payment.
• Factors outside our reasonable control, such as force majeure or failures of networks we do not operate.

7. Hardware replacement (dedicated servers)

For dedicated servers, we aim to replace failed hardware within 4 hours of diagnosis. This is a target, not a guarantee: diagnosis time and parts availability vary by location and component. We keep you updated in your client area while we work.

8. Support response targets

We aim to send an initial reply within 12 hours of a support ticket. This is a target rather than a guaranteed response time. Urgent matters, such as active abuse or a network-wide incident, are prioritized. Tickets are opened from the Support page of the client area and are available to customers with an active dedicated or GPU server; replies appear in the client area, never by email. Every other service is self-service: the Server actions on its page are carried out by our team. To send something sensitive, encrypt it with our PGP key and paste the armored text in your ticket.

9. Scheduled maintenance

From time to time we perform maintenance that may briefly affect a service. We announce scheduled maintenance at least 48 hours in advance, normally on the status page and, where relevant, in your client area, and we aim to schedule it during low-traffic periods. Maintenance announced this way is excluded from downtime under section 6. Rarely, emergency maintenance may be required at shorter notice to protect the network or your data; we give as much notice as circumstances allow. You can follow current and planned work on the status page, which is updated by our team.