According to the latest reports, Microsoft Azure virtual device service today is facing up to 6 hours of interruption. Between 05:12 UTC and 11:45 UTC on October 13, 2021, a subset of Azure customers using Windows virtual devices faced problems when performing service management operations such as launching, creating, updating, and deleting.
The deployment and update of the new virtual device also failed. Linux-based virtual devices and existing Windows virtual devices are not affected by this issue. In addition, when creating resources, services that depend on Windows virtual devices may also experience similar failures.
Microsoft announced the cause of the failure:
We found that the call failed during the service management operation because the required artifact version data could not be queried. Our investigation focused on the back-end computing resource provider (CRP) to determine the reason for the call failure and determine The required VMGuestAgent cannot be queried from the repository.
The VM Guest Agent Extension publishing architecture is being migrated (as part of the migration of the traditional service management back-end system) to a new platform that utilizes the latest Azure Resource Manager (ARM) Function.
Mitigation measures. We mitigate the impact by marking the appropriate extension as the correct expected level (in this case public). The engineers proactively verified that after the update was completed, the operation returned to a complete success rate.
The next steps. We will continue to investigate to determine the complete root cause and prevent future occurrences. The complete root cause analysis (RCA) will be announced within 72 hours.