Register / Data centre / EIR-1990
← Back to the list
Data centreUnited StatesEIR-1990

Reported power outage at Equinix Ashburn data centre linked to AWS connectivity problems

Mar 2018Published 4 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): Amazon Web Services, Equinix

On Friday 2 March 2018, a nor'easter storm knocked out utility power in the Washington region, including Loudoun County, Virginia. Equinix said that the weather affected power at a property it leases in Ashburn, Virginia, and that it activated its contingency plans, but these did not prevent a service interruption for some customers. Amazon Web Services reported that its Direct Connect service had lost connectivity to Equinix data centres DC1 to DC6 and DC10 to DC12 on the Ashburn campus, and to CoreSite facilities VA1 and VA2 in nearby Reston.

Direct Connect issues began around 6:20 am Eastern and lasted close to four hours for some customers. CoreSite said its facilities did not go down in terms of uptime, and did not explain why Direct Connect links to them had dropped. Equinix said the affected facility was operating on normal utility power again. The source did not establish whether the outage affected the whole Ashburn campus or a single facility housing shared Direct Connect infrastructure. The Equinix attribution rests on one source.

Regional utility power loss caused by a nor'easter storm. The source does not confirm the full cause of the failure of the contingency arrangements, or whether backup power was unavailable or failed to start. A single facility may have housed critical infrastructure serving all Direct Connect links on the campus, but this was not confirmed.

  1. Confirm that backup power and transfer systems at colocation sites carry the full critical load, and test them under real utility loss conditions.
  2. Map which shared network or interconnection equipment sits in which building, so a single-building power loss does not silently take down links for many customers.
  3. Customers relying on a single private cloud link should set up redundant links with fast failure detection, as the cloud provider itself recommended.

More lessons on outages, backup supply and restoration, detection and monitoring, from every incident on the register.

Site page: Equinix (status, dates and other incidents)

Other entries that name Equinix

Published 2026-10-04. 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 in the United States, by year, free with a credit (PNG).

Data centre incidents in the United States: 49 entries, with a free alert for the next one.

Browse the full register or submit an incident.