Fujitsu has acknowledged that a major power transformer failure took down its Sunnyvale data center on August 22nd. The failure was not within the data center itself but in the substation that provides power to the facility and under the control of their energy provider, not Fujitsu.
Besides highlighting the need for N+1 power provisioning for critical data centers, it appears that the external failure did cause cascading failures within the data center proper. Fujitsu announced that it was working to minimize the impact of the power outage on its customer base and mitigating the problems caused but there has been no announcement as to the actual problems that the failure caused in the data center or what was done to prevent a recurrence.
At least one customer reported that their Fujitsu cloud services were down for no less than five days after the original incident. This extended outage indicates some sort of failure in the data center's graceful shutdown strategy. A simple external power failure should have thrown the data center on to its backup power supply, where procedures should have been in place to gracefully shutdown systems without losing user data or damaging applications if it was determined that primary power could not be restored within the functional range of the UPS systems.
As most commercial data centers maintain UPS systems and generator backups that can provide power for an indefinite period of time should there be sufficient fuel, the problems with customer services being restored indicate either significant cascading damage being caused by the external transformer problem or a problem with the internal procedures within the Fujitsu data center for dealing with power outages.