On a Friday morning, as Britain began its working week, a large Microsoft outage that had already run for roughly twelve hours was still shutting people out of the digital tools they rely on to speak, write and work. Teams, Outlook and Copilot were failing for users across the world, and the restorations engineers had managed overnight had given way again. It is the familiar reminder that much of modern work rests on machinery few of its users ever see, and that when it falters, the fault is felt at once in ordinary offices and homes. What remains unsettled is whether the cause will be named a
Microsoft outage enters second half-day as Teams, Outlook and Copilot remain down after failed fixes
Recovery had come, and then slipped away
On a Friday morning, as Britain began its working week, a large Microsoft outage that had already run for roughly twelve hours was still shutting people out of the digital tools they rely on to speak, write and work. Teams, Outlook and Copilot were failing for users across the world, and the restorations engineers had managed overnight had given way again. It is the familiar reminder that much of modern work rests on machinery few of its users ever see, and that when it falters, the fault is felt at once in ordinary offices and homes. What remains unsettled is whether the cause will be named and whether recovery will hold.
By the time the UK woke up on Friday, the outage had already been running for about twelve hours. Microsoft Teams, Outlook and the Copilot AI assistant were still failing for users around the world, and engineers had not yet found a fix that held. Workers in Britain began their day unable to reach the tools many of them depend on first thing.
In a Friday morning update, Microsoft thanked affected users for their patience and acknowledged that serious problems continued. The company said its servers had been brought back during the night, UK time, only to crash again. It had isolated parts of its infrastructure and shifted traffic onto alternate routes, but neither step settled things, and it confirmed that a change it had applied was not taking effect as expected.
The source gives no root cause, no count of affected users and no restoration time. Reports of trouble at OpenAI, the company behind ChatGPT, arrived in the same period, though nothing so far connects the two. Attention now turns to whether Microsoft names a cause and whether any recovery lasts.
By the time the UK woke up on Friday, the problem had already been running for about 12 hours. Microsoft Teams, Outlook and the Copilot AI assistant were still failing for users around the world, and the company's engineers had not yet found a fix that held.
The outage hit several platforms in the Microsoft suite at once: email through Outlook, workplace messaging through Teams, and Copilot. Users in Britain began the working day to find they could not get into the tools many of them depend on first thing. OpenAI, the company behind ChatGPT, also reported problems, though the source does not say how or whether those were connected to Microsoft's.
In an update on Friday morning, Microsoft said it appreciated the patience of the people affected. It also acknowledged that serious problems were continuing. The company said it had managed to get its servers running more than once during the night, UK time, but those servers had then crashed again. Recovery had come, in other words, and then slipped away.
Microsoft described several attempts to contain the failure. It had isolated parts of its infrastructure, and it had shifted traffic onto alternate routes. Neither step settled things. The company said it had watched the status of the changes it applied to improve serviceability and confirmed that the change was not taking effect as expected.
That statement is the clearest account so far of where the repair effort stood. The source gives no root cause, no estimate of how many users were affected, and no time by which service would be restored. Microsoft's own wording, a change not taking as expected, points to fixes that were applied but did not do what engineers intended.
What to watch next is whether Microsoft names a cause and whether any restoration lasts rather than collapsing again as it did overnight. The reports of OpenAI trouble in the same period will also draw attention to whether the two companies' services share infrastructure, though nothing in the reporting confirms that.
Citazioni salienti
The change is not taking as expected.— Microsoft, in a status update
Vuoi la storia completa?
Leggi l'originale su The Mirror ↗