Updated May 26 with comment from Equinix, cause.
A major power failure at Equinix's aging MA1 data centre in Manchester this morning dropped ISPs and other customers off a cliff briefly this morning before most scrambled to initiate their failover mechanisms.
The MA1 data centre ncident affects the Internet Exchange, Metro Connect, Equinix Fabric and Equinix Connect services running out of its MA - Manchester location, the co-location data centre provider confirmed.
"The issue is related to IBX incident 5-217434657711. Power load was transferred to generators. IBX is still investigating issue and working to resolve it" Equinix said in a status update at 11:36am GMT+1.
IBX refers to Equinix's own network of 220+ International Business Exchange data centers.
MA1, first opened in 1998, is in Manchester Science Park and features low-mid tier N+1 power redundancy (i.e. power components, N, have at least one independent backup component. The Uptime Institute defines four tiers of data centre resilience: N, N+1, 2N and 2N+1, where N is the base and 2N+1 is the most resilient.)
"We had multiple customers on different providers and different areas of the UK go down for about 10 minutes at around 10:10 this morning" one observer noted on the /sysadmin sub-Reddit; Broadband and IT services provider Broadwell was among those affected, telling customers it had suffered a 6 minute drop in connectivity.
Equinix MA1 has been described as "the most important network node outside of London in the United Kingdom" and was bought by Equinix from Telecity in 2015. It is both small and somewhat dated however.
The site offers ~15,000 square feet of technical capacity at 5Amps/1kW per rack. Some 84 out of just over 100 LINX ports in use in Manchester are located in MA1 according to Colo-X.
Earlier this year Equinix said it was closing the neighbouring MA2 data centre in Manchester in June 2023 when the lease on the property expires, telling customers the facility no longer met its evaluation criteria.
Equinix told The Stack: "The incident occurred following planned maintenance work on the UPS (uninterrupted power supply) system, to replace one of the communications boards. The resulting power outage was rectified after a few milliseconds, with some services impacted for up to ten minutes as they rebooted. The operations team and relevant vendors are continuing to investigate the trip, to avoid similar incidents going forward."
The incident follows a 5-hour power outage at Interxion's LON1 data centre in January that took down trading on the London Metals Exchange as well as numerous other customers. Throughout that outage the company kept silent, later telling customers that customer support systems were taken out with the outage.