A fire in one building was enough to degrade service across an entire region
On a Tuesday in Delhi, a fire at a third-party data centre forced Google Cloud into an emergency shutdown of its networking equipment, sending disruptions cascading across India's digital infrastructure. Cities as distant as Chennai and Mumbai felt the slowdown, as businesses discovered that the cloud — for all its vastness — is still anchored to physical buildings that can burn. The incident is a quiet reminder that the most modern systems remain tethered to the oldest vulnerabilities: fire, power, and the fragility of the physical world.
- A fire at a Delhi data centre triggered an automatic power cutoff, severing a critical network hub and plunging Google Cloud customers across India into unreliable service.
- Delhi, Chennai, Mumbai, and surrounding regions all suffered elevated latency, turning apps sluggish, stalling websites, and straining the cloud-dependent internal networks of thousands of businesses.
- Google Cloud acknowledged on its status page that no workaround existed — leaving startups, enterprises, and AI-dependent operations with no escape route but to wait.
- The outage exposed a structural vulnerability: Google Cloud's reliance on third-party facilities means a single building fire can cascade into a regional digital crisis beyond the company's direct control.
- Google said it was exploring additional traffic mitigation measures, but offered no timeline for full restoration, leaving the scale of the disruption — and any injuries or property damage — undisclosed.
A fire at a third-party data centre in Delhi forced Google Cloud to shut down networking equipment on an emergency basis Tuesday, cutting off a critical hub serving the Indian capital and degrading connectivity across the metropolitan region. The blaze triggered an automatic power cutoff that isolated local infrastructure, and the effects spread quickly.
Network traffic from Delhi, Chennai, Mumbai, and surrounding areas all experienced elevated latency. Apps slowed, websites stalled, and company networks straining against cloud dependency began to buckle. Google Cloud — one of the world's largest providers alongside Amazon Web Services and Microsoft Azure — did not immediately disclose when the fire started, whether anyone was injured, or the extent of property damage.
For affected businesses, there was no workaround. Google acknowledged as much on its status page, leaving thousands of organizations — from startups to large enterprises running AI workloads — with no option but to wait out the restoration. The company said it was exploring additional traffic mitigation measures but gave no timeline for full recovery.
The incident laid bare a quiet structural risk: major cloud providers often rely on third-party facilities to house networking equipment in key markets. When one of those buildings catches fire, the failures cascade far beyond what any software fix can address. A single building in Delhi proved enough to degrade digital infrastructure across an entire region — a reminder that even the largest cloud systems are ultimately grounded in the physical world.
A fire at a third-party data centre in Delhi forced Google Cloud to shut down networking equipment on an emergency basis Tuesday, leaving customers across India without reliable service. The blaze triggered an automatic power cutoff at the facility, which isolated a critical network hub serving the capital and degraded connectivity throughout the metropolitan area.
The outage rippled across multiple Indian cities. Network traffic from Delhi, Chennai, Mumbai, and surrounding regions all experienced periods of elevated latency—the kind of slowdown that compounds quickly across dependent systems. Apps became sluggish. Websites stalled. Internal company networks that rely on cloud infrastructure to function began to strain under the load. Google Cloud, which competes with Amazon Web Services and Microsoft Azure as one of the world's largest cloud providers, did not immediately disclose when the fire started, whether anyone was injured, or what property damage occurred.
For businesses running on Google Cloud infrastructure, the disruption offered no escape hatch. The company acknowledged on its status page that no workaround existed while restoration efforts continued. Thousands of organizations across India—from startups to enterprises processing massive data volumes or running artificial intelligence tools—found themselves waiting for the company to restore capacity.
The incident exposed a vulnerability in how cloud infrastructure depends on physical facilities beyond a provider's direct control. Google Cloud relies on third-party data centres to house some of its networking equipment and points of presence in key markets. When one of those facilities catches fire, the cascade of failures can affect millions of users and businesses with no immediate alternative. The company said it was exploring additional traffic mitigation measures to limit future impact, but offered no timeline for when service would fully normalize.
The outage serves as a reminder that even the largest cloud providers operate within physical constraints. A fire in one building in Delhi was enough to degrade service across an entire region of India, affecting the digital infrastructure that modern business now depends on to function.
Citações Notáveis
Google Cloud said it was exploring additional traffic mitigation measures to limit the impact— Google Cloud status page
A Conversa do Hearth Outra perspectiva sobre a história
Why does a fire at one data centre affect so many cities at once?
Because that facility housed a critical network hub—a point of presence that routes traffic for the entire region. When it went offline, all the traffic that normally flows through Delhi had nowhere to go, so it backed up and slowed down across Mumbai, Chennai, and everywhere else connected to that same infrastructure.
Could Google Cloud have prevented this?
Not the fire itself, but possibly the cascade. They rely on third-party facilities they don't directly control. A more distributed architecture, or redundancy built in from the start, might have meant traffic could reroute automatically instead of just degrading.
What does "no workaround" mean for a business running on Google Cloud?
It means you're stuck. Your app is slow or down. Your data processing stalls. You can't switch to another provider mid-outage. You wait.
How long did this last?
The source doesn't say. That's actually telling—Google Cloud didn't disclose when the fire happened or when service fully restored, which suggests either they were still working on it or they wanted to move past it quickly.
Is this a common problem?
Not common, but not unprecedented. Any infrastructure that depends on physical facilities has this risk. A fire, a power failure, a cable cut—these things happen. The question is whether you've built enough redundancy to survive them.