Microsoft’s services remain down worldwide after more than 12 hours

It’s chaos. I can guess what’s happened. My educated guess, but if I were a betting man I reckon it’ll be a cascade failure to do with the health check, some configuration change internally that triggered it, and progressive inability of other servers to cope with the extra load i.e. a cascading failure. If that’s what’s happened, they will be frantically trying to balance the loads across their servers and find out what triggered the initial health check failure.

“It’s very fluid as Microsoft are not saying great deal. However, what it shows is how dependent so many are on the Microsoft server estate working properly. It has all sorts of knock on effects such as when a system sends you a code to log in via your email.

“And I don’t believe this is just email. I think it’s significant parts if there M365 application suite.

When things like this Microsoft outage happen, people immediately start looking for workarounds just to keep things moving, and that means switching to tools and systems that aren't the ones normally used or approved by organisations.

Organisations need to be careful with what people are doing now, when their usual systems are down, and what tools and programs people are using, as that introduces a potential risk. The other concern is what happens when systems come back. 

Organisations should consider having approved back-up systems in place and a level of redundancy to ensure people can switch to alternative systems and programs without risk.

The outage affecting Outlook and other major Microsoft services like Microsoft Teams is incredibly widespread, and its ramifications extend well beyond being able to send and receive work emails. Outlook and Teams are critical global business infrastructure. This outage is making it difficult for many organisations to properly field customer service requests received over the phone, as well as disabling their email communications. 

It highlights the dependence of Australian business and government on global technology providers like Microsoft. It also highlights the inherent dangers of having so many companies and organisations reliant on a very small number of tech firms. If one of those firms suffers a major outage, it has the potential to dramatically reduce business productivity globally during the outage time.

At this stage, there is no evidence that this is a cyber attack. We have seen from past outages like those involving Optus or those caused by CrowdStrike that widespread outages can result from simple configuration changes, even in the absence of malicious activity. 

The Microsoft outage appears to have been caused by a misconfiguration of networking systems and infrastructure. The impact has been global and highlights the risks associated with relying on cloud services. The loss of productivity will be in the tens of millions in Australia and tens of billions globally.

It is incumbent on the Australian government to ensure that big tech isolates Australia from outages of this type by having infrastructure and systems hosted locally. There should be a mechanism for Australian businesses to claim compensation and for the Government to fine big tech for outages that impact productivity.

The failure of Microsoft to ensure that best engineering practices were adhered to identifies how important it is for businesses to have tech risk management plans and alternatives in place to minimise the cost of tech outages.