SHD / MC Checker

[serviceRestored] EX1056704 | Exchange Online | Users may be unable to run Outlook (classic) in online mode and admins unable to utilize the Exchange admin center (EAC)



EX1056704 | Exchange Online | Users may be unable to run Outlook (classic) in online mode and admins unable to utilize the Exchange admin center (EAC)

Status serviceRestored
Classification advisory
User Impact Users may have been unable to run Outlook (classic) in online mode and admins may have been unable to utilize the EAC.
Last Updated 04/29/2025 17:23:37
Start Time 04/14/2025 00:00:00
End Time 04/27/2025 23:30:00
Latest Message Title: Users may be unable to run Outlook (classic) in online mode and admins unable to utilize the Exchange admin center (EAC)

User impact: Users may have been unable to run Outlook (classic) in online mode and admins may have been unable to utilize the EAC.

More info: Some users may have received the following error: “HTTP 503 Service Unavailable.”

Additionally, when attempting to complete operations in the EAC, admins may have been returned a “503 Service Unavailable” error. This could potentially have impacted any EAC related operation, including but not limited to mailbox provisioning. Admins who could leverage PowerShell to complete these tasks could’ve done so to avoid this issue.

Final status: We’ve confirmed from service telemetry that this issue is no longer occurring after reducing the retry traffic, and we’ve determined that this issue is resolved.

Scope of impact: Any user who attempted to start Outlook (classic) in online mode or any admin utilizing the EAC may have been impacted by this event.

Start time: Monday, April 14, 2025, at 12:00 AM UTC

End time: Sunday, April 27, 2025, at 11:30 PM UTC

Root cause: A high number of app requests and retries to an API that facilitates Outlook (classic) and the EAC were resulting in the impact.

Next steps:
– We’re continuing our investigation to better understand the underlying cause of the higher-than-normal rate of app requests and retries on the affected API to prevent issues like this from occurring in the future.

This is the final update for the event.