If we needed reminding of how precarious our increasingly fragile digital existence has become, Big Tech is happy to provide it.
On the surface, it sounds so innocuous: an inadvertent configuration change that was the “trigger event” for the issue.
But that one change, however it happened, hit Microsoft’s Azure services this week, specifically Azure Front Door. While you may never have heard of it, that service acts as an entry point for web applications built on the company’s servers. And the resulting “issue” meant companies all over the world were negatively impacted.
Delays in Microsoft’s Outlook email system and problems accessing Microsoft 365 online services were reported. Airlines, including Air New Zealand, found customers could not check in, causing delays in airports. The Scottish parliament’s voting system had technical problems, causing a debate to be abandoned.
RM Block
Numerous websites went down, including Asda in the UK and coffee giant Starbucks. Even children and gamers were discommoded as it affected Microsoft’s Xbox games platform, including Minecraft.
There were uncomfortable, albeit brief, echoes of last July’s Microsoft outage caused by a faulty security software update that caused widespread chaos for days afterwards.
Within a few hours, Microsoft had rolled back the change, and services were restored. Embarrassingly, the latest Microsoft problem came only hours before the company was due to report its latest financial results. But the damage that the incident has inflicted on people’s confidence in cloud services will be harder to repair.
It does not help that the latest cloud service outage is the second in as many weeks. Amazon’s AWS suffered a serious outage recently that disrupted apps and websites ranging from cryptocurrency exchange Coinbase to AI search engine Perplexity. It even affected the UK tax authority and Amazon’s own services. Requests to Alexa went unanswered, and Ring doorbell owners reported gaps in footage.
It was yet another reminder that we have become increasingly dependent on a small number of companies to keep our digital worlds operating. Cloud computing is largely dominated by three big companies: Microsoft, Amazon and Google, with other major players that include IBM, Oracle and Alibaba.
These companies have big investment budgets and access to data centre capacity, giving customers a cost-effective way to access cloud capacity to support their businesses. But the knock-on effect is that if one provider has a problem, the impact is widespread.
Even if companies choose to go to other cloud providers for their services, it is likely that somewhere along the way a component will rely on at least one of the Big Tech companies in some capacity. It is worth noting that last July’s CrowdStrike outage was caused by a faulty update pushed by the security provider to its customers, affecting more than eight million machines.
In other words, a disruption in one service could impact companies that aren’t even direct customers, because somewhere along the line, they are linked.
Microsoft suffered an “inadvertent configuration change”. In doing so, it reminded us once more of how much at the mercy we are of companies never making a mistake.
