Skip to main content
ENGLISH
Deutsch

Service Level Agreement

System availability

System availability is measured in terms of uptime over one month. Uptime is measured by an independent third party (UptimeRobot) and is defined as ability to access the web application and the shared blockchain API.

The guaranteed monthly uptime is 99.9%.

This uptime guarantee covers the following web applications:

This includes all blockchains listed as supported by our API on the main website:

There are additional blockchains available like zCash but we do not provide an uptime guarantee for those (yet).

Exceptions

Downtime caused by any of the following is not included when measuring uptime.

  1. General internet issues, including congestion or general DNS services not directly relating to Chain49 or services provided by Chain49
  2. Blockchain downtime, as in the whole network is down without our fault (Example: Solana had multiple complete downtimes in the past)
  3. Force majeure events, defined as unforeseeable events outside of the parties' control, that renders performance impossible or impractical
  4. Actions or inactions of the Customer (unless undertaken at the express direction of Chain49) or third parties beyond the control of Chain49.
    Note: Subcontractors of Chain49 are not considered a third party and Chain49 takes full responsibility for subcontractors
  5. A result of Customer equipment or third party computer hardware, software, or network infrastructure not within the sole control of Chain49
  6. Scheduled SaaS infrastructure maintenance
Compensation

If uptime is lower than guaranteed in a certain month, the Customer will be compensated with a reduced price per user in that month. Compensated Downtime is calculated as the maximum of (guaranteed uptime - actual uptime) across all of the sites.

Prices will be reduced with 5% for every 0.1 percentage point of Compensated Downtime. If for example Bitcoin has uptime of 100%, Ethereum has uptime of 99.6% and Dogecoin has up time of 99.8%, the compensated downtime will be maximum of (99.9%-99.6%=0.3%) and (99.9%-99.8%=0.1%), which is 0.3%. This results in a price reduction of 15% for that month (0.3% / 0.1% = 3, 3 * 5% = 15%).

The maximum price reduction will be 100%.

Reporting

Current and historical uptime statistics can be found on our status page at status.chain49.com.

It's hosted by UptimeRobot and we can not remove incidents from the history because we do not control the database.

Incident management

Incidents are addressed within 24 hours (within office hours).

Problems are categorised and handled as described here:

  • DEFCON 5: A minor problem that does not severely affect day-to-day business for the Customer.
    With an available workaround.

    Service target: This is scheduled by Chain49 based on feedback from the Customer and considering the inconvenience of the necessary workaround.

  • DEFCON 4: A minor problem that does not severely affect the Customer's ability to use Chain49 effectively.
    No available workaround.

    Service target: This is scheduled upon identification in collaboration with the Customer.
    It is determined on a case by case basis, but normally is scheduled for the next development cycle.

  • DEFCON 3: A problem that affects the Customer's ability to use Chain49.

    Service target: This is scheduled upon identification in collaboration with the Customer.
    It is determined on a case by case basis, but normally is scheduled for the next development cycle.

  • DEFCON 2: A problem that causes a feature or single chain to be inaccessible.

    Service target: This is solved in the current development cycle, or sooner.
    Max. resolution time: 7 days. Expected resolution time: 2-3 days.

  • DEFCON 1: A problem that causes Chain49 to be inaccessible.

    Service target: Resolution begins immediately even if outside office hours and continues until solved.


Backup & contingency plans

All data is backed up in three levels, making sure that substantial data loss is not possible.

  1. For essential data like API keys, single entities are logged directly in the database in order to be able to restore accidentally deleted data or for revision tracking.
  2. The complete application servers, including OS, are backed up nightly, providing an extremely fast and dependable method of restoring normal operations in case of catastrophic partial failure.
  3. In case of a catastrophic failure, i.e. flooding in the data centre or other factors causing severe hardware malfunction, rendering the main hosting centre unoperational, Chain49 has remote nightly backups. These can be used to restore normal operations. Chain49 has a secondary data centre, which can be used in case the primary data centre is not operational within reasonable time.

Upon catastrophic failure, the upper maximum limit for data loss is 24 hours. This is extremely unlikely. Procedures for catastrophic failure is simulated and tested annually.


Click the following button to download this document as a PDF file. If you do not have a device that can read PDFs by default, you can download a free program to read PDFs here: PDFCreator.