Google Cloud disruption in India after fire at third-party data center prompts traffic reroutes

Alphabet’s Google Cloud said some customers in India faced intermittent network disruptions after a fire at a third-party data center facility, according to reports citing Google Cloud updates. The disruption, centered around connectivity capacity in the Delhi metropolitan area, also appeared to affect traffic originating from other major cities, as Google rerouted portions of network traffic to maintain service.

Multiple outlets reported that the incident began late Tuesday night after emergency power shutdown procedures were triggered at the facility. Affected capacity was linked to a local Point of Presence, or POP, which is a geographic location used to connect network traffic to the broader internet. Reduced capacity at that POP prompted Google to divert a significant volume of traffic away from the site, creating uneven performance conditions for some users.

The reports described symptoms that customers may observe during the incident window. Google Cloud indicated customers could experience elevated latency, meaning slower response times, and possible packet loss, where some data packets do not reach their destination. The outlets said users in and around Delhi, Mumbai, and Chennai reported periodic connectivity issues consistent with those network performance changes.

One report said Google was still working on restoration and that it had not provided a customer workaround at the time of publication. In practical terms, that means affected customers could not simply switch to an alternative setup on their own side while Google continued efforts to restore the impacted network capacity.

The incident highlights how cloud service reliability can depend not only on a provider’s own infrastructure, but also on third-party facilities that supply network capacity and connectivity. Even when a major cloud platform can reroute traffic, reduced capacity in a critical region can cascade into performance issues elsewhere, especially when demand is spread across multiple metropolitan internet routes.

For Alphabet, Google Cloud is an increasingly important part of its business, and reliability expectations are tight because customers use cloud networks for online applications, data processing, and time-sensitive workloads. Disruptions like this also tend to raise questions about incident communications, transparency on restoration timelines, and how quickly performance normalizes once a fire-related shutdown ends.

Still, the publicly available reporting does not provide several details that would normally matter for a fuller risk assessment. The outlets did not cite confirmed information about the extent of physical damage, whether injuries occurred, or a precise timetable for when services returned to pre-incident performance. It is also unclear how many customers were directly impacted, or whether the disruption affected all Google Cloud services equally or only certain connectivity paths.

What to watch next is whether Google Cloud posts an incident resolution update showing restoration progress and final service status, and whether subsequent communications specify the scope of impact and the factors that drove traffic rerouting. Investors and enterprise customers will be watching for evidence that performance stabilized without recurring packet-loss patterns as the network returns to normal capacity. Meanwhile, similar incidents in shared or third-party facilities remain a reminder that even hyperscale networks can be constrained by localized outages.