Service level agreement
Availability, how it is measured, planned maintenance, backups and availability discounts.
Last updated: September 1, 2026 · In effect from: September 1, 2026
Download as PDFArticle 1 - Purpose and scope
1.1 This agreement (the “SLA”) sets the level of availability of the Services, how it is measured, and the compensation where it is not met. It forms an integral part of the Contract under Article 3.2 of the Terms.
1.2 The SLA applies to: web hosting, WordPress hosting, reseller hosting, e-mail hosting, hosted apps, virtual servers (VPS), Raspberry hosting and dedicated servers.
1.3 For Raspberry hosting, instead of the guaranteed level of availability under Article 3.1 the special regime under Articles 3.3 and 3.4 applies.
1.4 The SLA does not apply to the registration and administration of domain names, to SSL certificates, or to services provided free of charge or on a trial basis.
1.5 Terms not defined here have the meaning given in the Terms.
Article 2 - Definitions
2.1 Availability means:
- a) for web hosting, WordPress hosting, reseller hosting and e-mail hosting, the state in which the server on which the Service runs responds at network level on the relevant protocol (HTTP/HTTPS for web, SMTP/IMAP/POP3 for e-mail);
- b) for virtual servers (VPS), Raspberry hosting, hosted apps and dedicated servers, the state in which the allocated machine or virtual server is reachable at network level from the Provider’s network.
2.2 Availability is not assessed by whether the Customer’s application works. A fault in the Customer’s code, configuration, database or extension is not an outage of the Service, even where it appears from the outside as an unavailable website. For Services under Article 2.1(b), the state of the operating system and of software installed by the Customer is likewise not assessed (Articles 26.1 and 28.3 of the Terms).
2.3 An outage is a continuous period during which the Service is not available, excluding the periods under Articles 4 and 5.
2.4 The assessment period is a calendar month.
2.5 Working hours means Monday to Friday from 9:00 to 17:00 CET, excluding days of rest and public holidays in the Czech Republic. A working day is a day on which working hours run.
2.6 An availability discount means a discount on the price of the Service concerned granted where the level of availability under Article 7 is not met. It is granted by set-off against the price of the next Billing Period and is not paid out in money; it is not a balance on the Customer’s account.
Article 3 - Level of availability
3.1 The Provider commits to availability of the Service of 99.9% per assessment period, which corresponds to roughly 40 to 45 minutes of outage per month depending on the length of the calendar month.
3.2 Availability is calculated as the proportion of time during which the Service was available to the total time of the assessment period, after deducting the periods under Articles 4 and 5.
3.3 Raspberry hosting runs on a single physical device dedicated to the Customer, which is not redundant. The Provider therefore does not guarantee the level of availability under Article 3.1 for it; instead it undertakes to replace a faulty device within 24 hours on working days under Article 2.5, running from the moment the fault is detected by its own monitoring or reported by the Customer, whichever comes first. Replacement means bringing the Service back into operation on a substitute device. Data from local storage is transferred only where the condition of the faulty device permits; backing it up is the Customer’s responsibility under Article 22.4 of the Terms.
3.4 For the time from detection of the fault to the Service being brought back into operation on a substitute device, the Customer is entitled to a pro-rata part of the monthly price of the Service concerned. It is claimed under Article 8 and granted by set-off under Article 7.4. The availability discount under Article 7 does not apply to Raspberry hosting.
Article 4 - Measurement
4.1 The Provider measures availability with its own monitoring system at a check interval of 1 minute.
4.2 An outage is a state established by two consecutive failed checks. The shortest recordable outage is therefore 2 minutes.
4.3 The data from the Provider’s monitoring is decisive. The Customer may submit its own records; if it disagrees with the Provider’s data, the parties will attempt to resolve the discrepancy by agreement.
4.4 For Services under Article 2.1(b) the Provider monitors the platform and the host nodes, not each allocated machine separately. An outage of such a Service is therefore counted from its detection by the Provider’s monitoring or from its report by the Customer, whichever comes first.
Article 5 - Planned maintenance
5.1 Planned maintenance is not counted as an outage.
5.2 The Provider announces it at least 48 hours in advance by e-mail or in the Customer Portal.
5.3 Maintenance normally takes place in the window from 00:00 to 06:00 CET and will not exceed 4 hours in an assessment period.
5.4 Maintenance necessary to avert an immediate threat to security or operations may be carried out by the Provider without prior notice; it will report it without undue delay.
Article 6 - Exclusions
6.1 Unavailability caused by the following is not treated as an outage:
- the Customer’s application, code, database or configuration,
- exhaustion of the resources allocated to the Service under Article 21 of the Terms,
- conduct of the Customer or of persons it has allowed to use the Service,
- suspension of the Service under the AUP or for non-payment of the price,
- force majeure,
- an outage on the Customer’s side or on the side of its connectivity provider,
- an outage at third parties outside the Provider’s control, in particular domain name registries and certification authorities,
- planned maintenance under Article 5,
- operation left running after the Contract has ended under Article 16.1 of the Terms - that is no longer the provision of the Service under the Contract.
6.2 Nor is unavailability caused by an attack whose scale exceeds the capacity of the Provider’s protective measures treated as an outage. Ordinary attacks against which the Provider does provide protection are not an exclusion.
Article 7 - Availability discount
7.1 If availability falls below the level under Article 3.1, the Customer becomes entitled, on a request under Article 8, to a discount on the monthly price of the Service concerned:
| Availability achieved | Discount |
|---|---|
| 99.0% to 99.9% | 10% |
| 95.0% to 99.0% | 25% |
| less than 95% | 50% |
7.2 The total of the discounts for an assessment period will not exceed 100% of the monthly price of the Service concerned.
7.3 Where the Service is prepaid for a longer period, the monthly price means the corresponding pro-rata part.
7.4 The discount is granted by set-off against the price of the next period, not by refund.
7.5 No entitlement to a discount arises for a period in which the Customer was in default on payment of the price.
Article 8 - Claiming the discount
8.1 The discount does not arise automatically. The Customer requests it through the Customer Portal or by e-mail no later than 30 days after the end of the assessment period.
8.2 In the request the Customer states the Service concerned and the time during which, on its findings, it was unavailable.
8.3 The Provider assesses the request and notifies the Customer within 30 days.
8.4 If the Customer does not claim the discount within the period under Article 8.1, the entitlement lapses.
Article 9 - Exclusivity of the remedy
9.1 The availability discount under Article 7, and for Raspberry hosting the pro-rata part of the price under Article 3.4, is the exclusive remedy of a Customer who is a Business Customer for a failure to meet the level of availability.
9.2 The rights of a Consumer arising from defective performance under the Civil Code are not affected by this Article.
Article 10 - Support response times
10.1 The Provider responds to the Customer’s requests as follows:
| Severity | Response time |
|---|---|
| Critical fault (Service unavailable) within working hours | within 2 hours |
| Critical fault outside working hours | within 4 hours |
| Other requests | within 8 working hours |
10.2 The response time means the time until the Provider’s first substantive reply, not the time until resolution. A request may be submitted at any time. For a critical fault the response time runs continuously, that is, outside working hours, at weekends and on days of rest as well. For other requests it runs during working hours under Article 2.5; where they are submitted outside those hours, it starts to run at the next start of working hours.
10.3 Failure to meet a response time gives rise to no entitlement to a discount under Article 7.
Article 11 - Backups
11.1 For shared hosting services (web hosting, WordPress hosting, reseller hosting, e-mail hosting) the Provider makes automatic backups daily, normally in the window from 01:00 to 05:00 CET.
11.2 What is backed up is the data needed to run the websites, applications and e-mail on the Service (Article 22.2 of the Terms), that is, in particular files, databases, mailboxes and scheduler (cron) jobs. Data unrelated to that operation is not covered by the backups (Article 22.6 of the Terms).
11.3 The last 14 daily restore points are retained.
11.4 Backups are stored on a physically separate server and are encrypted.
11.5 Backups are not geographically separated. They are held in the same data centre as the production servers and therefore do not protect against the failure or destruction of the whole site.
11.6 Within the scope under Article 11.2 there is no cap on the size of the account or of the database for backup purposes.
11.7 The Provider does not back up VPS, Raspberry hosting, hosted apps or dedicated servers. Backing up their content is entirely the Customer’s responsibility. Network storage provided with the Service is space for the Customer’s own backups, not backups performed by the Provider.
11.8 Backups are an additional service provided without warranty under Article 22 of the Terms. They do not replace the Customer’s own backups and the Customer must not rely on them as its only source of data recovery.
Article 12 - Restoring from a backup
12.1 The Customer may carry out a restore itself in cPanel, at any time and free of charge.
12.2 On request the Provider will carry out the restore, likewise free of charge, within 24 hours on working days under Article 2.5.
12.3 A restore overwrites the current data within the scope restored. Before submitting a request the Customer should consider whether to download the current state first.
12.4 After the Contract ends the Content is made available for download for 14 days under Article 16.1 of the Terms. On a request made within that period the Provider will make the Content available within 24 hours on working days under Article 2.5. After that period the Content is retained until the 30th day from the end of the Contract under Article 16.2 of the Terms and can be made available only by the means set out in that paragraph; the period under Article 2.5 does not apply to them. On the expiry of the 30th day the Content is permanently deleted. If the Customer restores the Service, this Article applies to backups created after the restoration; backups from the period of the original Contract may no longer exist, because the retention under Article 22.2 of the Terms continues to run after it ends.
12.5 This Article does not apply to data that is not covered by the backups under Article 22.6 of the Terms.
Article 13 - Changes to the SLA
13.1 Changes are governed by Article 40 of the Terms.
13.2 A change that means for the Customer a reduction in the level of availability, in the discounts under Article 7 or in the scope of backups will be notified by the Provider at least 30 days in advance, and within that period the Customer may terminate the Contract.