Latest / Decrypted: The UK Cyber Briefing / A Microsoft outage showed how one network change can stop your email
Transcript
- 0:03Good morning, and welcome to Decrypted.
- 0:05I'm Belinder. It's Monday, the sixteenth of November, twenty twenty-six, and this is the morning edition.
- 0:12A Microsoft outage showed how one network change can stop your email.
- 0:17For more than a day this week, a big part of the working world couldn't rely on its email.
- 0:23From Thursday evening, UK time, Exchange Online, the Microsoft-run mail service behind Outlook, stopped working
- 0:30for customers worldwide. Teams, Copilot and Defender for Office three-six-five went with it.
- 0:36Microsoft says it was not an attack.
- 0:39That's the interesting part.
- 0:41Nobody needed to break in for a global outage to happen.
- 0:44So what actually went wrong?
- 0:47Microsoft's account, as it emerged through status updates, is a chain.
- 0:51An earlier network capacity upgrade left a fault.
- 0:54It surfaced after a later network change.
- 0:57Servers began failing their automated health checks and dropped out.
- 1:01The remaining servers took the extra load, and a core process on them became overwhelmed too.
- 1:07Microsoft said it was a technical issue not related to cyber or AI agents, and that it found no sign of malicious
- 1:14activity. Picture a motorway where one lane closes.
- 1:17Traffic floods the others, they jam in turn, and soon the whole road is stationary.
- 1:23Restarts helped, but The Register reported that some recovered servers could become unresponsive again once traffic
- 1:29returned. That's the jam re-forming as cars rejoin.
- 1:34And here's the awkward bit: the tools for fixing it were on that same road.
- 1:39Administrators also lost access to Intune and the Microsoft three-six-five Apps Admin Center, the tools used
- 1:45to manage devices and software, according to The Register.
- 1:48The controls for responding shared the fate of the thing they were meant to fix.
- 1:53Microsoft extended its email timeout window to 48 hours.
- 1:56That gives queued mail longer to arrive, but it also means a backlog that drains slowly.
- 2:02Dame Chi Onwurah, chair of the Commons Science and Technology Committee, said the outage "only serves to highlight
- 2:09once again the lack of resilience in the UK's digital infrastructure".
- 2:13That's a fair reading. Many UK organisations have quietly made one supplier the single route for staff, customers
- 2:20and administrators alike. So what's the lesson from Secure by Design?
- 2:25It's not only about keeping attackers out.
- 2:28It asks what happens when a component fails, and whether the failure stays small.
- 2:33Three decisions on the customer side would have limited this one.
- 2:36First, keep a second way to reach staff and customers that shares no supplier with your email.
- 2:41A phone tree, or a messaging service on separate infrastructure.
- 2:45Second, hold a tested break-glass route for administration, so losing one console doesn't leave you blind.
- 2:52Third, know which processes silently depend on a mailbox, from password resets to supplier invoices.
- 2:59None of this is expensive.
- 3:01Most of it is a rehearsal and a one-page list.
- 3:04The costly version is learning it live, on a Friday, with clients ringing.
- 3:09Microsoft has promised a full post-incident report once its review is complete.
- 3:13Boards should ask for it, and read it for the root cause rather than the apology.
- 3:18As always, this is opinion, not advice.
- 3:21That was Decrypted, with me, Belinder.
- 3:26The written version and every source are linked in the show notes.
- 3:29Thanks for listening. Stay patched, and stay sceptical.