The details of the outage, which lasted a week, were shared in a joint statement released by UniSuper and Google on May 8. Initially attributed to a “combination of rare issues at Google Cloud,” further explanations clarified that an “inadvertent misconfiguration” was to blame, ultimately resulting in the deletion of UniSuper’s entire Private Cloud subscription.
Thomas Kurian, CEO of Google Cloud, referred to the incident as a “one-of-a-kind occurrence,” reassuring stakeholders that measures are in place to prevent a recurrence. “We are deeply committed to ensuring this does not happen again,” Kurian stated, emphasizing their focus on bolstering protocols.
Recovery and Restoration Efforts
The loss of the Private Cloud subscription was particularly disruptive as it wiped out essential data across two geographic locations, designed to safeguard against such failures. Fortunately, UniSuper had backups with another service provider, which proved critical in the recovery process. “Restoring UniSuper’s private cloud instance has required an incredible amount of focus, effort, and partnership between our teams,” the joint statement highlighted.
This intensive collaboration has been central to the extensive recovery of all core systems affected by the outage. As of May 13, UniSuper reported that most of its online services had been restored, allowing members to access their accounts, make transactions, and utilize most features. “We’re still working on restoring our retirement calculators, but we expect to have these operational soon,” the company assured its users.
This is something I’ve not year heard before. Australian trading service UniSuper is down for the second day because of… Google Cloud.
Their outage was caused by a Google Cloud issue, as confirmed by the GC team.
GC meanwhile provides no details beyond “we’re really sorry!” pic.twitter.com/i4yuoSQQuQ
— Gergely Orosz (@GergelyOrosz) May 8, 2024
UniSuper’s Cloud Migration and Previous Challenges
The recent outage underscores the risks associated with cloud migrations, which UniSuper had been aggressively pursuing. In 2023, UniSuper had moved a large portion of its operations, including 1,900 virtual machines, to the Google Cloud Platform.
This transition was part of a broader strategy to streamline their infrastructure by moving away from Azure and their own data centers. UniSuper is no stranger to operational disruptions. A fire in 2017 at its Port Melbourne facility previously challenged the fund’s infrastructure resilience. Such incidents have shaped UniSuper’s approach to data management and system backups.
Broader Implications for Cloud Services
This incident raises important questions about the reliability of cloud services and the potential consequences of their failures. As businesses increasingly rely on cloud solutions for critical operations, the need for robust safeguards and swift recovery strategies becomes paramount. The fallout from this outage is a crucial reminder of the potential impacts on customer trust and business operations, pushing cloud service providers like Google to further refine their security measures and response strategies.
UniSuper’s Google Cloud Incident: Lessons Learned
UniSuper’s ordeal with Google Cloud serves as a case study in the challenges and risks inherent in digital transformations and cloud dependency. For the finance and tech industries, this event is a call to evaluate and enhance their infrastructure to prevent similar incidents in the future.
As for Google Cloud, the incident is a stark reminder of the responsibilities they bear and the expectations from their clients for reliability and rapid response in crisis management.
