Service Level Agreement
Effective Date: July 10, 2026
This Service Level Agreement ("SLA") applies to dedicated server, colocation, and cloud services provided by MinRTT Inc and forms part of our Terms of Service. It does not apply to optional add-ons or one-time services unless explicitly stated in writing.
1. Network Uptime
We commit to 100% network availability for services with redundant uplinks (see eligibility below), measured monthly, calculated as:
(Total minutes in month − Eligible Downtime) / Total minutes in month
"Network availability" means the ability to reach our edge routers from the public internet. Availability is measured from the perspective of our upstream transit providers, not from a single customer location.
Eligibility — redundant uplinks required. This network uptime commitment applies only to services configured with redundant uplinks (for example, LACP-bonded server connectivity across two switches, or a colocation cabinet provisioned with redundant uplinks to separate switches or routers). Services connected by a single, non-redundant uplink — including any dedicated server or colocation cabinet that uses or elects a single network port — are not covered by this network uptime commitment, because a single uplink has no failover path and a switch, port, optic, or link fault will interrupt connectivity with no redundancy to absorb it. Customers who require a network uptime guarantee must select a redundant-uplink configuration.
Dedicated servers are provisioned with redundant, LACP-bonded connectivity across two switches by default. If a customer reconfigures a service to a non-redundant state — for example by breaking the bond, disabling or failing to negotiate LACP, forcing a single link, or operating over a single NIC or port — the network uptime commitment does not apply for any period the service remains non-redundant, even though the redundant physical links are present and available.
2. Power Availability (Colocation)
For colocation customers, each cabinet is served by two independent power feeds (A and B) through MinRTT's redundant PDUs. We commit that at least one of the two PDU feeds will be available at all times, with a 99.9% monthly availability commitment calculated on the same basis as Section 1.
Power Eligible Downtime means any period during which both the A and B feeds to the cabinet are simultaneously unavailable due to a failure of MinRTT's PDUs or the power MinRTT delivers to them, and counts as Eligible Downtime for the service credit calculation in Section 7. The loss of a single feed while the other remains available is not downtime, because the redundant feed is designed to carry the full load.
This commitment covers only MinRTT's PDUs and the power delivered to them. It does not cover the customer's own equipment, power supplies (PSUs), power cords, internal cabling, or any failure to connect equipment across both feeds. To benefit from this commitment, the customer must connect redundant PSUs across both the A and B feeds; equipment connected to only one feed is not covered.
3. Higher Availability Tiers
Our network is engineered for — and backed by — a 100% uptime SLA: dual-homed transit, LACP-bonded server connectivity across independent switches, redundant power, and 24/7 hardware response. A single server is still a single machine, however: hardware can fail and an operating system reboot takes a server offline even while the network around it stays up, which is why the 100% commitment in Section 1 covers the network, not the individual server.
If that is not good enough (for example, regulated workloads with strict availability requirements, global low-latency applications, or true geographic redundancy), we offer 100% uptime solutions built on redundant anycast architectures spanning multiple data centers. These deployments are fully managed by our team and include active-active failover, BGP-controlled traffic steering, and continuous health monitoring across all sites. Pricing starts at $2,500/month for a complete solution: two fully managed servers across two data centers with a dedicated /24 anycast announcement. Contact sales@minrtt.com to discuss requirements.
4. Hardware Replacement
For dedicated server and colocation customers, we will begin diagnosis of suspected hardware faults within 2 hours of a support ticket being opened, and replace confirmed failed components within 4 hours of confirmation, 24/7. Replacement applies to drives, memory, power supplies, NICs, and similar field-replaceable parts. Whole-chassis replacement may take longer depending on spare availability and will be communicated case-by-case.
This applies only to hardware we own. Customer-owned colocation hardware is the customer's responsibility; we will provide remote hands at standard rates.
5. Scheduled Maintenance
We schedule maintenance during low-usage windows (00:00–06:00 US Central Time) where possible. Standard maintenance is announced at least 72 hours in advance via email and the control panel. Emergency maintenance (security patches, urgent hardware intervention) may be performed without notice when delay would put customer data, service stability, or the network at greater risk.
Time spent in announced or emergency maintenance does not count as downtime.
6. Exclusions
The following are excluded from uptime calculations and do not qualify for service credits:
- DDoS attacks and null routes. Any period during which a customer IP address, device, or service is the target of a denial-of-service attack, is receiving attack traffic, or is null-routed — whether the null route is applied by MinRTT or by an upstream carrier, and as described in our AUP — is excluded from Eligible Downtime. This includes reasonable time to confirm an attack has subsided before a null route is removed. Customers requiring uptime during attacks should purchase our paid DDoS protection add-on or our anycast tier described in Section 3.
- Issues caused by the customer's own software, operating system, configuration, firewall rules, or applications.
- Issues caused by third-party services or networks outside our control, including upstream transit failures we have already engineered around (we are multi-homed; single-carrier failures should not be customer-visible, but multi-carrier failures may be).
- Scheduled maintenance announced under Section 5.
- Emergency maintenance under Section 5.
- Force majeure: natural disasters, war, terrorism, government action, utility failures beyond our facility, and similar events outside our reasonable control.
- Suspension or termination for breach of the Terms of Service or AUP.
- Failures of customer hardware (in colocation) or customer-installed components.
- Failure of customer power supplies (PSUs), power cords, internal cabling, or equipment; loss of a single power feed while the other remains available; or any cabinet whose equipment is not connected across both the A and B feeds. Power availability is covered only under Section 2, and only when both feeds fail simultaneously.
- Services configured with a single, non-redundant uplink, or whose default redundant configuration has been disabled, broken, or removed by the customer, which are not eligible for the network uptime commitment in Section 1.
- Any period during which the customer has unpaid invoices more than 14 days past due.
7. Service Credits
If we fail to meet the network (100%) or power (99.9%) uptime commitment in a calendar month, eligible customers may request a service credit equal to five times the prorated value of the downtime:
Credit = 5 × (Eligible Downtime in minutes / Total minutes in month) × Monthly fee for affected service
For the 100% network commitment, there is no downtime allowance: all Eligible Downtime in the month counts toward the credit calculation from the first minute.
For the 99.9% power commitment, 99.9% corresponds to approximately 43 minutes of allowed downtime per 30-day month. If Power Eligible Downtime stays within that allowance, no credit applies; once exceeded, all Power Eligible Downtime for that month counts toward the credit calculation.
Worked examples for network downtime (assuming a $1,000/month service, 30-day month, 43,200 total minutes):
| Eligible Downtime | Credit |
|---|---|
| 15 minutes | $1.74 |
| 1 hour | $6.94 |
| 4 hours | $27.78 |
| 12 hours | $83.33 |
| 24 hours | $166.67 |
| 6 days or more | $1,000 (capped) |
For power downtime the same table applies, except that no credit accrues unless Power Eligible Downtime for the month exceeds the 43-minute allowance.
If we miss a hardware replacement commitment under Section 4, the period beyond the stated response time counts as Eligible Downtime for the purpose of this calculation, credited from the first minute beyond the commitment.
Credits are capped at 100% of the monthly fee for the affected service in the month the downtime occurred, regardless of cause or number of incidents. Credits are applied to the next invoice, are not redeemable for cash, and do not extend beyond the current month of service.
8. Claim Process
To request a service credit:
- Open a support ticket within 30 days of the end of the affected month.
- Include the affected service ID, the dates and times of the incident(s), and any reference numbers from our status updates.
- We will review and respond within 10 business days. If we agree the SLA was missed and the incident is not excluded under Section 6, the credit will appear on your next invoice.
Failure to submit a claim within 30 days waives the right to credit for that month.
9. Sole Remedy
Service credits under this SLA are your sole and exclusive remedy for any failure to meet these service levels. Our total liability for any SLA failure in any given month is limited to the monthly fee paid for the affected service for that month. This SLA does not expand the limitations of liability in the Terms of Service.
10. Changes
We may update this SLA with 30 days' notice. Changes apply prospectively only and do not affect credits already accrued.
Contact: Submit a support ticket / sales@minrtt.com / +1.844.818.2656 / 14425 Falcon Head Blvd BUILDING E, Suite 100, Austin, TX 78738.