Report on Large-Scale System Failure in the Keio University Network (Jan. 27) (Interim Recovery Completed)

(Published: Jan. 27, 2026)

Earlier today (January 27), a large-scale system failure occurred due to a power facility malfunction at the data center housing Keio University's server groups. We sincerely apologize for the significant inconvenience caused to all faculty, staff, and students. As of 6:50 PM, we have confirmed the startup of major servers and hereby issue an interim declaration of recovery.

According to ongoing reports from the data center operator, the current restoration has been achieved through temporary power sources. Consequently, there is a possibility that services may be suspended again when switching back to the main power supply, depending on the facility configuration and technical requirements. We will continue to coordinate and respond to this situation with the highest priority on minimizing the impact on university operations and events.

13:55  Detected anomalies, such as inability to log in to keio.jp.
Initial assessment suggested a large-scale failure at the data center housing the virtualization infrastructure.
14:05  Confirmed that an error occurred at the data center (Cause under investigation).
14:07  Began internal preliminary announcements. (At 14:55, provided updates regarding the scope of the outage and active services including KIF4, K-Support, and Slack).
14:20  Identified the cause as a "power failure" at the data center.
15:55  Received notification that power within the facility was restored. Started verifying the recovery of university systems and services.
16:10  Physically inspected the leased racks at the data center; confirmed no abnormalities in LEDs, exhaust, or operational sounds.
16:50  Official Keio University website restored.
18:50  Issued the overall recovery report for today, including notices regarding future response.

Updates will be provided on this page as they become available.