All systems are operational
Maintenance
Data center migration - services intermittently unavailable

Because of a migration of a significant part of SNTs servers to a different data center, our services will intermittently be unavailable throughout the entire day. This (indirectly) affects almost all internal and external services. This includes our mirrors, IRC servers, mail relay, GitLab CI runners, and Matrix(-related services).

There is no expected downtime for vColo, and websites hosted by SNT in general. An exception to this is the Hornet application itself. This means that, as Hornet will have limited availability throughout the day, you might not be able to log into the administration panel for your websites/domains hosted by SNT. We apologize for the inconvenience.

Past Incidents

2021-04-19

GitLab security update

GitLab has released a security update. Due to its severity SNT's instance will be upgraded immediately.

2020-12-21

Database outage

The central database server is experiencing a major outage. Investigation is ongoing.

  • Issue was resolved with manual intervention. All affected services are back online.

    • IRC bridges Matrix outage

    All Matrix-related services are experiencing an outage because of database issues.

  • The major outage has been resolved. Because of ongoing maintenance there will be some performance issues, and probably some service interruptions.

    • 2020-11-03

    GitLab GitLab security update

    GitLab has released a security update. Due to its severity SNT's instance will be upgraded immediately.

  • The update completed successfully

    • 2020-09-10

    GitLab security update

    GitLab has released a security update. Due to its severity SNT's instance will be upgraded immediately.

  • After the upgrade, it became apparent that newer versions of both Ruby and git were needed. This set back the completion of the update by a few hours.

    Everything should work as intended now.

  • A major version upgrade was required to fix the standing security issues, which required an unexpected database upgrade. This took half an hour extra.