Microsoft 365 caught a bug and the whole world sneezed

While Microsoft seems at last to be on the edge of fixing the global outage that hammered its services overnight, causing chaos across Asia-Pacific and Europe, we should still ask: How many times must we endure outages of this kind? How long will it be until government, enterprise, and everyday users understand the need to rely on hybrid systems to drive our digital age, rather than placing complete trust in a single provider?

It is, after all, not so very long since the CrowdStrike disaster impacted business users everywhere. This time round we are enduring a massive Microsoft 365 outage that is affecting people across the planet. The problem is that with so many systems reliant on Microsoft’s products, the ripple effect of failure is pressing into adjacent sectors. Police, health services, governments, and all kinds of businesses are facing disruption.

One vendor, many casualties

Why are they facing this? Because they rely on a single vendor to provide the back end for much of their tech. Even entities you’d imagine wouldn’t be impacted are feeling the pressure. Anthropic has revealed issues with its services. In the UK, the NHS reports that while clinical and operational systems are functioning, communication systems are impacted. At time of writing, the Bank of England has confirmed internal communications issues. All of this chaos is caused by an overreliance on Microsoft.

Even new employee onboarding will be impacted, as the outage impacts Intune’s ability to handle device enrolment, check-in, and compliance. Plus, of course, every third-party system that relies on core Microsoft systems has also failed.

It doesn’t need to be this way. Alternatives to Microsoft do exist, and while they may not have captured the business zeitgeist to the same extent as Outlook or Office, in the context of today’s global service disaster, it makes sense for every business to invest more in those alternatives. 

“This should be a wake-up call for every organisation. Businesses need to ensure they can continue operating when critical technology becomes unavailable, without relying entirely on a single system or provider,” said Marijus Briedis, Chief Technology Officer at NordVPN, in a statement seen by Computerworld. “Secure backup communications, tested contingency plans, and robust recovery procedures can make the difference between a temporary inconvenience and a major operational crisis.”

How long is too long?

What’s really surprising about the outage is how long it has taken for Microsoft to fix the problem. It is clearly trying to do so, but its status page messages aren’t terribly clear, and the fixes seem far from smooth. It would almost be better to think the company was being hacked than that it was struggling to fix its own mistakes, but Microsoft has not claimed to be hacked. Experts believe these global problems have been caused by a software update that contained some kind of “bug.”

A bug that made the world sneeze.

The best way to avoid such bugs is to build in reliance on multiple systems. At the front end, that means deployment of macOS, ChromeOS, Linux, and iOS. At the back end, it means investment in new infrastructure. 

The outage is also wind beneath the wings of arguments for digital sovereignty. UK politician Dawn Butler wrote: “The global Outlook outage is a reminder of why digital sovereignty matters. We need a Parliamentary, London and UK wide strategy to invest in our own capability and take control of our digital future. We cannot rely on US Big Tech for our digital infrastructure, including AI.”

The sovereignty question

That may be true, but while we wait for local tech developers to build systems that are at least as good while also being more secure than those offered by the US tech giants, we’re going to need to use something, and once again the best approach is continued investment in diversity, backup, and resilience. 

That’s going to mean more enterprise purchasers making the very sensible decision to invest in Macs, iPhones, and other non-Windows devices — which may not protect against cloud service outages, but may well support alternative services to boost disaster resilience. It’s going to demand investment in backup communication systems, alternate email providers, and Slack, and it’s going to be a very good reason indeed for IT to find alternatives to Intune for device management. 

It’s going to mean those things because where we are right now, they are the only available options, other than the magical thinking that some new entity is going to form to offer viable, secure alternatives to current operating system vendors. 

What happens Monday?

In the short term, I anticipate business users will start next week to build new secondary communication and email channels, create emergency admin accounts using offline credentials, invoke new backup routines for M365 data, and revise their outage policies. That’s when they finish reading their M365 SLAs to determine if there are financial consequences to the company’s latest global outage.

Now please subscribe to my daily, human-curated Apple-related news headline feed at The Core, or follow me on BlueSky, LinkedIn, or Mastodon.ComputerworldRead More