Posts / microsoft
The Quiet Joy of a Fifteen-Hour Email Outage
Somewhere in the middle of the weekend-adjacent Microsoft 365 outage, a user on Reddit wrote that the situation was inefficient because they could still do their tasks without Copilot and Outlook. They asked Microsoft to please make it worse. It was marked as a joke, but it didn’t read like one.
The outage ran past fifteen hours. Each fix failed in its own interesting way. By the time Europe was heading home and the US was waking up, the thread had settled into a mood I recognise well: grim relief. People were swapping tactics for not receiving email. Go offline. Point at the news articles. Say you assumed Teams was down too. One person said HR was meant to send them a termination notice, and funnily enough it never arrived.
I laughed at that. I also felt a bit sad about it.
I work in IT, so I’m on the other side of these days. When Exchange Online falls over, nobody in my world is delighted. Someone is in a call at 2am refreshing a status page that is, as usual, green. There is a particular kind of helplessness in being responsible for a system you don’t run. You can’t fix it. You can only tell people it’s not your fault, and nobody wants to hear that.
That’s the part I keep returning to. Most organisations handed their email, calendar, files, chat and now their AI assistant to one vendor, and called it simplification. It is simpler, right up until it isn’t. Then everything fails together, and the failure is somebody else’s schedule.
The other thread that caught me was about fax machines. Japanese offices apparently dusted theirs off. Someone pointed out that the FAA requires fax capability, partly for exactly this reason. Someone else described a government office where certain agencies have no public email at all, so documents move by fax or post. Their IT staff found this baffling. Another person tried to fax a partner and ended up talking a stranger through what a fax machine was.
I’m not about to argue for fax. But there’s a lesson in the clunky redundancy. The old way was dull, and it was also boring to break. Nobody pushed a config change that took out every fax machine on the planet at once.
Resilience is unglamorous and it doesn’t demo well. You can’t put “we kept a second way of doing it” in a slide about efficiency. I’ve lost that argument in budget meetings, and I expect to lose it again. Paying for a spare feels like waste right up until the day it’s the only thing working.
There’s a tension I can’t resolve. I like the convenience. I use a pile of cloud services, and I’m not going to run my own mail server in the spare room again. Anyone who did that in 2008, as I did, remembers it with a wince. Big vendors are usually more reliable than I’d be. They also fail bigger. Both things are true, and I don’t have a tidy answer for which wins.
What I noticed in the thread, under the jokes, was how many people were quietly glad to be freed. A day without email turned out to be a day when some work got done, and plenty of it moved into Teams chat instead, which is a funny kind of rebellion. People pasted entire email threads into chat, which is the same problem in a different box.
Anyway. If you’re responsible for any of this, pick one thing this week that has no second path and write down what you’d do if it vanished. It doesn’t need to be a project. A page will do. When the status page stays green and the phones start ringing, you’ll be glad it exists.