Microsoft 365 had two separate problems on 5 October 2026: SharePoint Online search stopped working for some users for 4h 38m, and the admin centre became intermittently unreachable for admins for 2h 58m. Both faults sat on Microsoft's side and both were posted to Microsoft's own status page.
Started
13:39 UTC
Duration
Lasted 4h 38m
Source
IsDown
Next time
Microsoft 365 is degraded. Is your product working?
An incident on their side does not always mean an outage on yours, and the only way to know is to already be watching your own endpoints. Set that up in under a minute and you will be notified the next time your own site is affected. And there will be a next time.
50 monitors for your own websites. No card.
Affected components: Microsoft 365 suite, SharePoint Online
What happened?
The first incident started at 13:39 UTC and affected the search box inside SharePoint Online. Microsoft's first update, at 13:40, said it was analysing information to work out next troubleshooting steps. By 15:51 Microsoft had narrowed this to a compatibility issue with a newly updated search service component, and said it was reviewing that component to find a root cause and a fix. That incident ran for 4h 38m before being marked resolved, though no closing note explaining the final fix appears in what Microsoft published. The second incident began at 22:28 UTC and hit the Microsoft 365 admin centre, with admins intermittently unable to get in. Microsoft's 23:38 update attributed this to an unusually large increase in requests to the admin centre, which it said was causing requests to queue and time out. By 00:08 Microsoft said it had implemented performance optimisations and validated that service availability had recovered to pre-impact levels, while it kept investigating the underlying cause. That second incident lasted 2h 58m. No user reports were filed in this window on either incident. Two components were affected across the day: SharePoint Online and the Microsoft 365 suite generally, via the admin centre. Microsoft's history shows 49 incidents in the last 90 days with a median duration of 7 hours 20 minutes, so both of these resolved faster than Microsoft's own recent median.
Learning
Neither of these faults was the kind of thing an external uptime check would catch on its own. A probe hitting a Microsoft 365 or SharePoint login page can come back green while the search box behind it fails to return results, or while admins specifically, not general users, are locked out of one console. That is a feature-level and role-level failure inside a provider's infrastructure, not a reachability failure at your edge, and your own endpoint monitoring has no visibility into Microsoft's search service or its admin centre request queue. UptimeRobot's own 30-day figures for this provider show 19 incidents with a median resolution time of 3h 16m, which is a useful baseline for how long these things typically take to clear. If your team was unaffected today it does not mean the provider was fine, it means the specific feature or role that broke was not one your workflow touches. Your monitoring covers your half, UptimeRobot now watches the provider's half too, so next time you know which side broke without guessing.
