Microsoft's 18-Hour Outage Exposed South Africa's Big Tech Dependence

An 18-hour Microsoft outage has disrupted some of South Africa's largest institutions, exposing how much of the country's economy runs on a handful of global technology providers.

Eskom, South Africa's power utility, warned that its email services had been hit by a worldwide disruption to Microsoft Exchange. FlySafair, the domestic airline, said it could temporarily neither send nor receive emails. Absa, one of South Africa's largest banks, reported intermittent difficulties accessing Microsoft's workplace applications.

The outage struck on Thursday and hit Microsoft's widely used productivity and cloud services, including Outlook, Exchange Online, Teams, OneDrive, SharePoint and Copilot. It stretched far beyond South Africa, affecting businesses, government institutions and educational organisations across multiple countries.

What did the affected companies say?

"Eskom's email services are currently impacted due to the worldwide disruption of the Microsoft Exchange service," the utility said in a Friday post on X. It directed customers to its telephone, WhatsApp, chatbot and USSD channels for assistance.

FlySafair warned customers that the disruption to Microsoft 365 had left it temporarily unable to send or receive messages. The airline advised passengers to use its call centre and WhatsApp instead.

Absa said the disruption was affecting aspects of its Microsoft environment, with some employees struggling to access applications including email and Copilot. "Microsoft is actively working on a resolution, and our tech teams are monitoring the situation closely," the bank said.

Other institutions reported problems too:

• The City of Cape Town said its Microsoft 365 environment, including email, Teams and OneDrive, had been affected.
• The South African Revenue Service, the country's national tax authority, issued an alert about the disruption to its services.

Why does one outage cut across so many sectors?

The three organisations operate in power, aviation and banking. Yet the same failure stopped their email, collaboration tools and customer-service channels at the same time. That is the mechanism of concentration risk: when a country's utilities, airlines, banks, municipalities and tax authority all run on one vendor's cloud, a single infrastructure failure spreads across industries and hits everything from internal communication to business continuity.

The consequences went beyond locked inboxes. Eskom had to reroute customers to alternative channels. FlySafair asked passengers to phone or message instead of emailing. Absa warned that staff tools were unavailable.

How did Microsoft respond?

Microsoft's recovery effort showed how hard it is to restart services that support millions of users. In updates on its service health channels, the company said it was investigating elevated server loads and unexpected server unavailability. Its attempts to redirect traffic away from affected infrastructure ran into further difficulties.

By Friday, Microsoft was reporting early signs of stabilisation and improvements in capacity as it worked to restart affected infrastructure.

"To help preserve delayed mail through this event, we've extended email timeout to 48 hours," the company said in an update on Friday morning.

The 48-hour timeout is a telling detail. Mail queued during the outage still needed two full days of buffer to survive the backlog — a measure of how deep the disruption ran.

What does this change for South African businesses?

The incident raises questions about how prepared South African businesses are for prolonged disruptions to the platforms underpinning their operations. Cloud services now sit at the centre of how organisations manage communications, store documents, collaborate and deliver services.

The outage forced Eskom, FlySafair and Absa to fall back on telephone lines and WhatsApp while Microsoft rebuilt its infrastructure. For businesses that rely on cloud platforms for daily operations, the episode made the case for backup systems and business continuity plans — before the next single-vendor failure repeats the pattern.