Workday Outage Locks Workers Out of Payroll and HR After Backup Test Goes Wrong
Workday, the cloud platform that runs payroll, timekeeping and human resources for more than 11,500 organisations worldwide, went down on Friday evening after a test of its own backup systems went wrong, leaving staff locked out of the tools they use to get paid, clock in and enrol in benefits.
Service was restored by 11 a.m. Eastern time on Saturday, a Workday spokesperson said, after an outage that stretched through the night and into the busiest part of the benefits-enrolment season. Some customers came back later, and universities reported delays in data syncing even after access returned.
Workday said only that “some” of its customers were affected, and it has not published a count of the organisations or employees caught up in the disruption. The company also confirmed the outage was not a cybersecurity incident.
What went wrong
The failure began while Workday was validating a weekly maintenance window for its disaster-recovery tenants — the backup environment designed to keep the service running if the main systems fail. During that test, services in the disaster-recovery site unexpectedly connected to the live production databases, leaving data in what Workday described as an inconsistent state.
A notice on Workday’s community status page, quoted by Washington University’s IT team, put it bluntly: “While validating the weekly Maintenance Window for Disaster Recovery tenants, we have identified that tenanted services within the DR Site unexpectedly connected to Production databases. As a result, this has left data in an inconsistent state that can cause errors accessing tenants or jobs/integrations/tasks to fail with missing transaction error messages.”
Workday placed its WD1 production tenants into maintenance and promised updates every 30 minutes while its engineers worked on a fix. The San Francisco Chronicle first reported that the problem had been found during backup testing, and a Workday spokesperson told local broadcasters: “Some of our customers are experiencing a service disruption and we’re working urgently to investigate and resolve the issue.”
The test existed to prove Workday could survive a failure. Instead, it helped create one.
Who felt it, and what stopped working
For affected staff, the outage was immediate: employees could not log in, clock in or out, check their pay or finish benefits enrolment, and some were in the middle of open enrolment, the once-a-year window for health cover choices.
Emerson College in Boston said Workday took the system offline at about 10.15 p.m. GMT on Friday for unplanned maintenance affecting multiple customers, and extended its open-enrolment deadline to 5 p.m. GMT on Monday, 16 November. Washington University was still listing a major outage in the early hours of Saturday, while Saint Joseph’s University said a potential solution was still being tested.
Outage tracker Downdetector logged nearly 3,000 problem reports shortly after 10 p.m. GMT on Friday, UC Today reported.
It is important to be precise about the scale. Workday’s customer list includes large retailers such as Walmart and Target, as well as Netflix, Salesforce and several US government agencies, but that is the size of its customer base, not a confirmed list of those affected. No source has confirmed that pay was not processed — the reported problem was access to payroll information and HR services, which is not the same as salaries going unpaid.
A second outage in one week
The Workday failure was the second major business-software outage in the space of a few days. Microsoft said on Saturday that Outlook, Teams and the rest of its Microsoft 365 services were fully restored after a global outage lasting roughly 36 hours, which the company traced to a fault left by a network capacity upgrade. That Microsoft 365 outage began on Thursday, knocked out email and calendars for users in more than 70 countries, and came in the same week as a nationwide FAA ground stop in the United States.
The run of failures has put a spotlight on how much of daily work now depends on a handful of cloud platforms. When one of them falters, the effects spread far beyond its own data centres: into payroll departments, university HR offices and the enrolment deadlines of workers who never chose the software in the first place.
What happens next
Workday told ABC10 that the disruption hit two data centres and that service “was restored by 11 a.m. EST Saturday”. Washington University marked the outage resolved later that afternoon, and the University of Chicago said Workday was operating normally by early evening, though some customers were still reporting limited recovery into Saturday night.
The questions now are the ones that follow every major outage. When will Workday publish a post-incident report? What do its service-level agreements credit customers for a night without payroll access? And, most pointedly, why did a disaster-recovery test have any path at all into live production data?
Until those answers arrive, the practical advice is a manual timekeeping fallback, a check of payroll and benefits deadlines, and a clear message that nobody will be penalised for a login they could not complete.
