If you asked a Magic 8-Ball about your Thursday afternoon, it would have said “Outlook not so good”. The Microsoft 365 outage that started on 12 November took Outlook, Teams and Exchange Online offline for people around the world, and for anyone whose working day lives in more than one calendar, it was a reminder that your availability is only as reliable as the systems holding it.
What happened in the Microsoft 365 outage
According to BleepingComputer’s report, the incident began at 1:26pm Eastern on Thursday. Microsoft’s advisory (MO1550821) warned that customers might be unable to connect to Exchange Online, Microsoft Teams, SharePoint Online, Microsoft 365 Copilot and Microsoft Defender XDR. Problems were first reported in the Americas, and Microsoft later confirmed that users in any region could be affected.
Microsoft first put it down to elevated load. It then found the load wasn’t coming from extra traffic at all: a number of servers had become unexpectedly unavailable, straining the ones that were left, and a core process on the affected servers was being overwhelmed. Microsoft paused all deployments, moved traffic away from the affected systems, restarted servers and looked at whether reverting a few recent configuration changes would speed things up. By Friday morning it was reporting “encouraging recovery progress”, and it had extended its email timeout window to 48 hours so that delayed messages wouldn’t be lost while the queues cleared.
Exchange Online also holds Outlook calendars. So for a few hours, a lot of people couldn’t see their meetings, couldn’t accept invites and couldn’t tell anyone when they were free.
Why it matters if you juggle several calendars
If all your meetings sit in one Microsoft 365 account, an outage is annoying but simple: everything stops, and everyone knows it. If you’re a consultant with a client’s Microsoft 365 account, your own Google Calendar and maybe a second client on top, it’s messier. One calendar goes dark while the others carry on taking bookings. The people booking you in Google can’t see the Outlook meeting they’re about to clash with, and you can’t check, because Outlook is down.
Then there’s the tail. With Microsoft holding mail for up to 48 hours, invites sent during the outage can land a day late, possibly on top of something you booked in the meantime.
What we learnt building Calendrz around other people’s outages
Calendrz does availability mirroring: when you’re busy in one calendar, it blocks that time in your others, without copying the private details. That means our product depends on Google and Microsoft being up, and we’ve had to design for the days when they aren’t.
A few of the decisions that came out of that:
- Every connected account is read on its own. If one provider fails, we record the error against that connected account and carry on with the rest of your Profile. One bad connection doesn’t stop the others being looked after.
- Not every failure is the same failure. A server error or a timeout from a provider is usually a blip, so we retry a couple of times with backoff and a little randomness, so thousands of retries don’t all land on a struggling service at once. An expired or revoked sign-in is different: retrying won’t fix it, so we don’t hammer away at it.
- We learnt some of this the hard way. A while ago, a single timeout from Google when writing a block showed up in our logs as an error, plus a duplicate warning, with no retry at all. Our founder, Liviu Tudor, changed that so writes like this are retried when the failure is temporary, and the problem is logged once, at a level that reflects whether anyone actually needs to act on it.
There’s an honest limit here too. When Microsoft is down, we can’t see what changed in your Microsoft 365 calendar either. Nobody can. What we can do is catch up properly once it’s back: the meetings that arrive late get auto-blocked in your other calendars as soon as we can read them.
What to take from it
- Check the outage window. Look over anything booked between Thursday afternoon and Friday morning, in every calendar, for clashes with invites that arrived late.
- Don’t let one provider be the only place your availability exists. If your Outlook meetings are also blocked out in your other calendars, people booking you there still see that you’re busy.
- Expect more of this, not less. Copilot went down with everything else. As more of the clicking and booking gets handed to AI assistants, an assistant is only as useful as the calendars it can reach, and a single point of failure becomes everyone’s problem at once.
Outages happen to the biggest platforms in the world. The fix isn’t to hope they stop; it’s to make sure one of them going quiet doesn’t leave you double-booked when it comes back.
If you juggle more than one calendar, try Calendrz free and let availability mirroring keep your calendars honest with each other, whichever one is having a bad day.







