Register / Data centre / EIR-1883
← Back to the list
Data centreBelgiumEIR-1883

Lightning strikes cause Google data centre power loss in Belgium

Aug 2015Published 2 Oct 2026Updated 13 Oct 2026Responsible editor: Chief Editor

Estimated damageEstimated by the register's adjusterHow? Know the real figure?
Figure hidden: shown to subscribersMedium confidence
out of service for about •• daysShown in the record →Full estimate
From the recordRoot cause hiddenFailed part hidden

Companies named in the reporting of this entry (being named is not a finding of fault): Google

At 09:19 PDT on 13 August 2015, four successive lightning strikes hit the local utilities grid supplying Google's European data centre in St. Ghislain, Belgium, within the europe-west1-b zone. The strikes caused a brief loss of power to storage systems hosting Google Compute Engine persistent disk capacity. Automatic auxiliary systems restored power quickly and the storage systems have battery backup, but some recently written data sat on storage systems more susceptible to power failure from extended or repeated battery drain.

Roughly 5% of Standard Persistent Disks in the zone had at least one I/O failure, and manual intervention was needed to return systems to normal serving state. Recovery continued until 17 August 2015, when a very small fraction of disk space, under 0.000001%, had permanent data loss. SSD Persistent Disks, snapshots and Local SSDs were not affected.

The storage systems that held recently written data were more susceptible to power failure from extended or repeated battery drain. Google's review identified several contributing factors across the hardware and software stack, which the company said it is working to address. Google stated that its own hardware upgrade programme was ongoing and that most Persistent Disk storage was already on less susceptible hardware.

  1. Repeated grid disturbances such as successive lightning strikes can exhaust battery backup, so backup endurance should be tested against repeated interruptions, not only a single outage.
  2. Identify which storage tiers hold recently written data that is most exposed to a power interruption, and prioritise those for hardware upgrades.
  3. Where data is critical, keep snapshots and geographically replicated copies, because a single facility or zone can fail as a whole.

More lessons on weather and external hazards, from every incident on the register.

Published 2026-10-02. Written from public reporting. Descriptive, not investigative. See the data accuracy disclaimer.

See the register's stories first in Google: add it as a preferred source.

Chart to reuse with this story: Data centre incidents by year, free with a credit (PNG).

Data centre incidents in Belgium: 2 entries, with a free alert for the next one.

See UPS and battery room fires at data centres.

Browse the full register or submit an incident.