Azure outage: Websites disabled in Microsoft global outage come back online

After encountering issues during a worldwide Microsoft outage, the websites for Heathrow, NatWest, and Minecraft went back online late on Wednesday.

Over the course of several hours, the outage tracker Downdetector displayed thousands of reports of problems with various websites worldwide.

According to Microsoft, some Microsoft 365 customers experienced delays with Outlook and other services. However, by 21:00GMT, the company had restored a previous update, and many of the downed websites were operational again.

At 16:00 GMT, a "degradation of some services" was reported by the company's Azure cloud computing platform, which powers a significant portion of the internet.

 The same "DNS issues" that caused last week's massive Amazon Web Services (AWS) outage were cited as the cause of this.

According to Amazon, AWS was functioning normally.

While people in the US reported having trouble accessing the websites of coffee chain Starbucks and retailer Kroger, other sites affected in the UK were supermarket Asda, M&S, and mobile phone operator O2.

Microsoft claimed that issues were encountered by business Microsoft 365 users.

Additionally, some Microsoft websites sent users to error messages that said, "Uh oh! The prior request had a problem.

After some users complained that they were unable to access the service status page, the tech giant turned to posting updates to a thread on X.

Although NatWest's website experienced a brief disruption, the bank's online chat, mobile banking, and phone customer support were all operational throughout the outage.

According to the UK consumer advocacy group Which?, companies had a duty to compensate affected customers and make sure they were informed and supported while services were restored.

Lisa Webb, a consumer law specialist with Which?, suggested that "customers should keep evidence of any failed or delayed payments in case they need to make a claim."

Ms. Webb continued, "Those who are concerned about missing a bill should get in touch with the appropriate business to explain the circumstance and ask that any fees be waived."

In the meantime, technical difficulties with the online voting system at the Scottish Parliament caused business to be suspended.

Debate over land reform laws that would enable Scotland to interfere in private sales and mandate the dissolution of large estates was put on hold as a result of the outage.

According to a senior Scottish Parliament source who spoke to BBC News, the issues were thought to be connected to the Microsoft outage.

Azure's vital function on the internet

Although it's unclear how much of the internet was affected, estimates generally place Microsoft Azure at 20% of the global cloud market.

According to the company, "an inadvertent configuration change" was the cause of the outage.

Put differently, there were unexpected repercussions from a change made to a system that operates behind the scenes.

An outage like this "can cripple hundreds, if not thousands of applications and systems," according to Dr. Saqib Kakvi of Royal Holloway University, because cloud services are concentrated in Microsoft, Amazon, and Google.

"Economic forces cause resources to be concentrated into a few very large players due to the cost of hosting web content, but this is essentially putting all of our eggs in one basket."
Gregory Falco, a professor of engineering at Cornell University, says recent outages have exposed the vulnerability of today's internet.

According to Mr. Falco, "When we think of Azure or AWS, we think of a monolithic piece of technology infrastructure, but the reality is that it's thousands, if not tens of thousands, of little pieces of a puzzle that are all interwoven together."

He pointed out that while third parties like CrowdStrike, which last year released a software update that impacted over eight million computers running Microsoft systems, are in charge of some of those pieces, the companies themselves are in charge of others.