Orbus Software experienced a minor incident on September 7, 2026 affecting OrbusInfinity, lasting 8h 3m. The incident has been resolved; the full update timeline is below.
Affected components
Update timeline
- identified Sep 07, 2026, 09:09 AM UTC
We have identified an issue affecting Power BI reports for customers on the OrbusInfinity UK South regional node. Customers attempting to load reports are receiving the following error: "Unable to load model due to reaching capacity limits" This is impacting Dashboards within OrbusInfinity also. Impact: Power BI reports are currently unavailable. Users are unable to open or load reports within the platform. Current Status: The cause has been identified. Our team is actively working on a resolution.
- identified Sep 07, 2026, 10:39 AM UTC
We have taken initial remediation steps, however these have not resolved the issue. Our engineers remain engaged and are continuing to investigate.
- monitoring Sep 07, 2026, 10:52 AM UTC
The issue affecting Power BI reports on the OrbusInfinity UK South regional node has been resolved. Services have been restored and will be monitored closely over the coming hours. Thank you for your patience.
- resolved Sep 07, 2026, 05:13 PM UTC
After a period of monitoring, we can confirm that the issue affecting Power BI reports on the OrbusInfinity UK South regional node has been fully resolved. Services are operating normally.
- postmortem Sep 11, 2026, 10:06 AM UTC
## Incident Summary On 05 September 2026 at 09:03 UTC, customers on the OrbusInfinity UK South regional node experienced an outage affecting Power BI reports and Dashboards. Users were unable to load reports, receiving a capacity limit error. The issue was identified on 07 September 2026 and resolved at 10:45 UTC on the same date, following which the service was monitored and confirmed fully stable at 17:13 UTC. ## Impact Customers on the OrbusInfinity UK South regional node were unable to access Power BI reports and Dashboards for the duration of the incident. **Regions Affected** * OrbusInfinity UK South **User Experience** * Power BI reports could not be loaded * OrbusInfinity Dashboards were unavailable * Users received the error: "Unable to load model due to reaching capacity limits" ## Root Cause Analysis The incident was caused by the expiry of an internal credential used by our automated capacity management service, which prevented the service from scaling compute capacity in response to increasing demand. **What Happened** In the early hours of 05 September 2026, our automated capacity management service scaled the UK South compute instance down to its base tier during a period of low utilisation. Shortly after, an internal credential used by the service expired, causing it to stop functioning. As demand increased through the day, the service was unable to scale capacity back up. CPU utilisation exceeded 100% at 09:03 UTC, at which point the capacity limit was reached and Power BI reports and Dashboards became unavailable. **Technical Details** * The automated Power BI capacity scaling service relies on an authenticated credential to interact with the Power BI API. This credential expired, causing the service to stop functioning. * The timing of the expiry, while the instance was already at its lowest capacity tier, meant the service could not respond to increasing demand, exacerbating the impact. * The manual elevation of capacity via our Cloud Orchestration portal failed to propagate as expected to the impacted reporting workspaces, resulting in further manual remediation activities being required to restore service. This behaviour is consistent with a Microsoft Power BI and Fabric platform issue \(Microsoft incident reference: PB1461484 - "Admins may see outdated/incomplete data in workspace settings of Power BI or Microsoft Fabric services", active 02 August 2026 to 02 September 2026\). While Microsoft reported this issue as resolved prior to our incident, the propagation failure observed is consistent with the same underlying platform behaviour, and it is possible that residual impact persisted in the UK South region beyond the stated resolution date. **Resolution** Compute capacity was manually elevated to resolve the immediate service impact. Further manual remediation was carried out within the Power BI administration portal to force the capacity change to propagate correctly to all workspaces, restoring service at 10:45 UTC on 07 September 2026. The expired credential was subsequently renewed, restoring full automated capacity management. The service was monitored closely and confirmed fully stable at 17:13 UTC. **Mitigations** As a result of this incident, the following improvements are being implemented to reduce the likelihood and impact of a recurrence: * Ownership of credential expiry notifications for the automated capacity management service has been expanded to ensure multiple teams receive advance warning ahead of expiry dates. * The automated capacity management service is being updated to send real-time failure notifications via Microsoft Teams, regardless of the cause of failure, to ensure issues are detected and acted upon immediately. * The automated capacity management service is being updated to dynamically handle new and rotated credentials, removing the dependency on manual intervention following a credential renewal. * Enhanced CPU alerting is being introduced for Power BI services to ensure that capacity degradation is detected and responded to proactively.