Service Level Agreement
Last updated:
We commit to 99.9% monthly network and server availability for hosting, VPS and email services. This document sets out how that is measured, what it excludes, and what you get if we miss it.
1. What’s covered
Availability of the network and of the host your service runs on: routing to your server, power and cooling at the facility, and the hypervisor or shared hosting platform itself.
2. How availability is measured
Availability is measured per calendar month, in whole minutes, from our external monitoring nodes in multiple locations. A minute counts as unavailable when the service fails to respond from a majority of nodes.
Monthly availability = (total minutes in the month − unavailable minutes) ÷ total minutes in the month.
3. Service credits
| Monthly availability | Credit |
|---|---|
| 99.9% or above | None — commitment met |
| 99.0% to 99.89% | 10% of that month’s fee |
| 95.0% to 98.99% | 25% of that month’s fee |
| 90.0% to 94.99% | 50% of that month’s fee |
| Below 90.0% | 100% of that month’s fee |
Credits are applied against future invoices for the affected service. They are not paid in cash, and total credits in any month cannot exceed that month’s fee for the affected service.
4. Claiming a credit
Open a ticket in the billing department within 30 days of the end of the affected month, stating the service, the dates and times of the outage, and any logs or third-party monitoring you have. We respond within five business days.
We do not apply credits automatically — you have to ask. Our monitoring is the reference record where there is a discrepancy, but we will look at yours.
5. Exclusions
Time is not counted as unavailable where downtime results from:
- Scheduled maintenance, announced at least 48 hours in advance and performed in a low-traffic window. We aim for under four hours per month.
- Emergency maintenance needed to address an active security threat, announced as far ahead as circumstances permit.
- Your application — code errors, exhausted PHP workers, a database brought down by a query, or a plugin that crashes the site.
- Exceeding plan resource limits, or throttling applied under the Acceptable Use Policy.
- Suspension for non-payment or policy breach.
- Your own changes — DNS misconfiguration, firewall rules, or actions taken with root access on a VPS.
- Factors outside our reasonable control — denial of service attacks against you, upstream provider failures, government action, or natural disaster.
6. Support response targets
| Severity | Target first response |
|---|---|
| Critical — service down | 15 minutes |
| High — significantly degraded | 1 hour |
| Normal — general technical issue | 4 hours |
| Low — question or feature request | 1 business day |
These are targets rather than contractual commitments; no service credit attaches to them. We publish our actual median response time on the status page.
7. Backups
We take daily off-site backups of hosting accounts with the retention stated on your plan, and restore on request. Backups are a disaster-recovery measure and not a contractual guarantee: you remain responsible for maintaining your own copies of anything you cannot afford to lose.
8. Sole remedy
Service credits under this agreement are your sole and exclusive remedy for any failure to meet the availability commitment, subject to the liability provisions of the Terms of Service.
9. Status and incidents
Current status, scheduled maintenance and post-incident reports are published at status.codexacloud.com. We publish a written post-mortem for any incident causing more than 30 minutes of downtime.