Orbus Software experienced a critical incident on September 15, 2026 affecting OrbusInfinity and OrbusInfinity, lasting 3h 4m. The incident has been resolved; the full update timeline is below.
Affected components
Update timeline
- investigating Sep 15, 2026, 02:01 PM UTC
We are currently investigating an issue affecting the OrbusInfinity platform in the South Africa and Qatar regions. Users in these regions may be experiencing a "Something went wrong" error when attempting to access the platform. Our engineers are engaged and actively investigating. We will provide a further update as soon as possible.
- monitoring Sep 15, 2026, 03:02 PM UTC
We can confirm that the OrbusInfinity platform in the South Africa and Qatar regions has been restored. We are actively monitoring both environments to ensure they remain stable.
- identified Sep 15, 2026, 03:29 PM UTC
The OrbusInfinity platform in the South Africa region has been restored and we are monitoring to ensure it remains stable. The Qatar region remains in a degraded state. Our engineers are continuing to investigate and work towards full restoration. We will provide a further update as soon as possible.
- monitoring Sep 15, 2026, 03:49 PM UTC
We can confirm that the OrbusInfinity platform in the South Africa and Qatar regions has been fully restored. We are actively monitoring both environments to ensure they remain stable.
- resolved Sep 15, 2026, 05:05 PM UTC
Following a period of stable monitoring, both the South Africa and Qatar OrbusInfinity nodes are healthy and stable. This incident is now resolved.
- postmortem Oct 16, 2026, 04:37 PM UTC
## Incident Summary On 15 September 2026, OrbusInfinity was unavailable to users in the South Africa and Qatar regions. Users who tried to access the platform received a "Something went wrong" error. The incident began at 13:56 UTC in South Africa and at 14:03 UTC in Qatar. It was caused by a preparatory configuration update, designed to have no impact on service, which overwrote a setting used by the application's health checks with an incorrect value. Service in South Africa was restored at 14:58 UTC and service in Qatar was restored at 15:39 UTC. After a period of monitoring, the incident was resolved at 17:05 UTC. ## Impact Users in the South Africa and Qatar regions could not access OrbusInfinity for the duration of the incident. Other regions were not affected. ### Regions Affected * South Africa * Qatar ### User Experience * Users received a "Something went wrong" error when trying to access OrbusInfinity. ### Regional Recovery Timeline | **Region** | **Incident Start \(UTC\)** | **Full Recovery \(UTC\)** | | --- | --- | --- | | South Africa | 13:56 | 14:58 | | Qatar | 14:03 | 15:39 | ## Root Cause Analysis On 15 September 2026, a preparatory configuration update was applied to the South Africa and Qatar regions ahead of planned maintenance to migrate them to a new hosting platform. The update was designed to have no impact on the live service, so it was not part of the advertised maintenance window. However, a misconfiguration in the automated deployment process meant that two stages of the process interpreted one of its parameters differently. As a result, the step that retrieves the existing configuration was skipped, but the step that writes configuration still ran, overwriting a setting used by the application's health checks with an incorrect value. The health checks then began to fail, even though the underlying platform services were operating normally. The platform's automated recovery took the application out of service and kept trying to recover it. While the incorrect setting remained in place, the application could not report a healthy status, so service could not be restored automatically. ### Resolution The correct configuration values were reinstated in both regions. Service was restored in South Africa at 14:58 UTC. Restoring the Qatar region needed a further configuration correction, and service there was restored at 15:39 UTC. Both regions were then monitored until the incident was resolved at 17:05 UTC. The planned maintenance scheduled for the following day was postponed while the cause was confirmed. The following changes have been made to address this issue: * The automated deployment process now interprets this parameter the same way at every stage. * The configuration step involved now includes an additional check before it runs. * The logic that caused this setting to be overwritten during migration preparation has been removed. These changes have been applied to the deployment process used for future migration preparation work.