Not again! Microsoft blames config tweak for 365 outage in parts of North America
(2025/08/21)
- Reference: 1755790211
- News link: https://www.theregister.co.uk/2025/08/21/microsoft_365_outage/
- Source link:
Microsoft had a midweek meltdown on Wednesday as a chunk of its productivity suite fell out of the cloud.
The problems kicked off around noon UK time, when users found themselves unable to access Office.com. The company [1]confirmed users' pain via its social media mouthpiece shortly afterwards.
A few hours later, Microsoft [2]said : "We've identified a recent change is contributing to impact." It pushed out a rollback.
[3]
After another hour or so, the company confirmed that the "reversion" of the change was complete and that the incident had been resolved. It [4]added : "Users may need to refresh their browsers to experience relief."
[5]
[6]
There was a little more detail in the Microsoft 365 Admin Center. According to the console, the incident lasted a little more than four hours and was limited to organizations hosted "within a specific section of affected infrastructure within the North America region."
According to Microsoft, the preliminary root cause was that "a recent configuration change resulted in errors when users attempted to access Office.com, leading to impact." Access to Copilot via m365.cloud.microsoft was also affected, and the company suggested using the Copilot for Microsoft 365 app or other 365 apps (such as Teams and Office) as a workaround.
[7]
Microsoft has not officially revealed what the configuration change that caused the chaos was, only that rolling it back fixed the issue.
The company has some history when it comes to configuration changes that wreak havoc on its cloud. Earlier this year, for example, it broke the web version of Outlook with a dodgy change. At the time, [8]we asked the question : "Does Microsoft test its changes before deploying to production?"
As similar self-inflicted incidents continue to cause problems for the company and its customers, we have to ask again: what testing is being conducted that a configuration change could disrupt the company's infrastructure in this way?
[9]FydeOS offers ChromeOS without the Google strings attached
[10]Microsoft stays mum about M365 Copilot on-demand security bypass
[11]Out-of-band update arrives to clean up Windows reset and recovery mess
[12]Latest Windows 11 insider builds hide secret File Explorer dark mode
The first of Microsoft's "Next steps" is "To help prevent similar impact in the future, we're further reviewing our testing and validation processes prior to deployment."
We asked Microsoft's chatbot, Copilot, what to do before deploying a configuration change to production. The assistant said: "Before deploying a configuration change to production, it's crucial to treat it with the same rigor as a code deployment. A misstep here can cause outages, security vulnerabilities, or performance degradation."
[13]
And yes, Copilot also recommended a bit of validation in a local or staging environment first.
It's good advice. But sadly, North American users might have been unable to access the assistant via Microsoft 365 due to the outage.
We asked Microsoft what the configuration change was, but the company has not responded. ®
Get our [14]Tech Resources
[1] https://x.com/MSFT365Status/status/1958160855649964376
[2] https://x.com/MSFT365Status/status/1958220791423639887
[3] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/applications&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=2&c=2aKdClz419fmMafz2_HPamwAAAAc&t=ct%3Dns%26unitnum%3D2%26raptor%3Dcondor%26pos%3Dtop%26test%3D0
[4] https://x.com/MSFT365Status/status/1958251830124716537
[5] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/applications&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44aKdClz419fmMafz2_HPamwAAAAc&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[6] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/applications&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33aKdClz419fmMafz2_HPamwAAAAc&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[7] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/applications&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44aKdClz419fmMafz2_HPamwAAAAc&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[8] https://www.theregister.com/2025/03/20/outlook_outage_again/
[9] https://www.theregister.com/2025/08/21/fydeos_chromiumos_degoogled/
[10] https://www.theregister.com/2025/08/20/microsoft_mum_about_m365_copilot/
[11] https://www.theregister.com/2025/08/20/microsoft_oob_reset_patch/
[12] https://www.theregister.com/2025/08/19/windows_11_insider_builds_file_explorer_dark_mode/
[13] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/applications&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33aKdClz419fmMafz2_HPamwAAAAc&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[14] https://whitepapers.theregister.com/
The problems kicked off around noon UK time, when users found themselves unable to access Office.com. The company [1]confirmed users' pain via its social media mouthpiece shortly afterwards.
A few hours later, Microsoft [2]said : "We've identified a recent change is contributing to impact." It pushed out a rollback.
[3]
After another hour or so, the company confirmed that the "reversion" of the change was complete and that the incident had been resolved. It [4]added : "Users may need to refresh their browsers to experience relief."
[5]
[6]
There was a little more detail in the Microsoft 365 Admin Center. According to the console, the incident lasted a little more than four hours and was limited to organizations hosted "within a specific section of affected infrastructure within the North America region."
According to Microsoft, the preliminary root cause was that "a recent configuration change resulted in errors when users attempted to access Office.com, leading to impact." Access to Copilot via m365.cloud.microsoft was also affected, and the company suggested using the Copilot for Microsoft 365 app or other 365 apps (such as Teams and Office) as a workaround.
[7]
Microsoft has not officially revealed what the configuration change that caused the chaos was, only that rolling it back fixed the issue.
The company has some history when it comes to configuration changes that wreak havoc on its cloud. Earlier this year, for example, it broke the web version of Outlook with a dodgy change. At the time, [8]we asked the question : "Does Microsoft test its changes before deploying to production?"
As similar self-inflicted incidents continue to cause problems for the company and its customers, we have to ask again: what testing is being conducted that a configuration change could disrupt the company's infrastructure in this way?
[9]FydeOS offers ChromeOS without the Google strings attached
[10]Microsoft stays mum about M365 Copilot on-demand security bypass
[11]Out-of-band update arrives to clean up Windows reset and recovery mess
[12]Latest Windows 11 insider builds hide secret File Explorer dark mode
The first of Microsoft's "Next steps" is "To help prevent similar impact in the future, we're further reviewing our testing and validation processes prior to deployment."
We asked Microsoft's chatbot, Copilot, what to do before deploying a configuration change to production. The assistant said: "Before deploying a configuration change to production, it's crucial to treat it with the same rigor as a code deployment. A misstep here can cause outages, security vulnerabilities, or performance degradation."
[13]
And yes, Copilot also recommended a bit of validation in a local or staging environment first.
It's good advice. But sadly, North American users might have been unable to access the assistant via Microsoft 365 due to the outage.
We asked Microsoft what the configuration change was, but the company has not responded. ®
Get our [14]Tech Resources
[1] https://x.com/MSFT365Status/status/1958160855649964376
[2] https://x.com/MSFT365Status/status/1958220791423639887
[3] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/applications&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=2&c=2aKdClz419fmMafz2_HPamwAAAAc&t=ct%3Dns%26unitnum%3D2%26raptor%3Dcondor%26pos%3Dtop%26test%3D0
[4] https://x.com/MSFT365Status/status/1958251830124716537
[5] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/applications&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44aKdClz419fmMafz2_HPamwAAAAc&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[6] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/applications&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33aKdClz419fmMafz2_HPamwAAAAc&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[7] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/applications&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44aKdClz419fmMafz2_HPamwAAAAc&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[8] https://www.theregister.com/2025/03/20/outlook_outage_again/
[9] https://www.theregister.com/2025/08/21/fydeos_chromiumos_degoogled/
[10] https://www.theregister.com/2025/08/20/microsoft_mum_about_m365_copilot/
[11] https://www.theregister.com/2025/08/20/microsoft_oob_reset_patch/
[12] https://www.theregister.com/2025/08/19/windows_11_insider_builds_file_explorer_dark_mode/
[13] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/applications&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33aKdClz419fmMafz2_HPamwAAAAc&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[14] https://whitepapers.theregister.com/
Doctor Syntax
"What testing is happening before changes hit production?"
But isn't production where it's tested?
QA. What's that?
"Users may need to refresh their browsers to experience relief."
More like.
Users may need to refresh their OS to experience relief.