A Microsoft Azure cloud platform outage on Wednesday, October 29, 2025, caused widespread disruptions across the Netherlands, knocking out NS's online journey planner, ticket machines, and OV-fiets bike rental system for several hours. Microsoft confirmed the incident stemmed from an accidental configuration change to its Azure Front Door service, which began around 16:00 UTC (5:00 p.m. local time).
What Happened
NS confirmed the disruption started shortly after 5:00 p.m. and linked it directly to the international Azure issue. The railway company emphasised that station information boards and access gates continued functioning normally throughout the incident: only digital ticket sales, the online journey planner, and bike rentals were affected.
Microsoft deployed its "last known good configuration" to restore services and began redirecting traffic to healthy infrastructure. By later in the evening, NS reported that systems were coming back online and ticket sales and bike rentals were resuming.
Wider Impact
The outage extended far beyond Dutch railways. Tens of thousands of users worldwide reported problems accessing Microsoft 365 applications including Outlook, Teams, and OneDrive, as well as consumer services like Xbox and Minecraft. In the Netherlands, outage monitoring services recorded complaint spikes for multiple Microsoft platforms, and other companies such as energy provider Vattenfall also experienced disruptions.
The Azure Front Door service eventually reached over 98% availability later in the evening, with Microsoft tracking toward full recovery by early morning on October 30.
Photo Credits: Lukas Boekhout/Unsplash
What Azure Front Door Does
Azure Front Door is a content delivery network (CDN) that speeds up how websites and applications deliver information to users. When it failed, many services that depend on Microsoft's cloud infrastructure experienced delays, timeouts, or complete unavailability.
Recovery Timeline
Microsoft blocked all new customer configuration changes at 17:30 UTC to prevent further impact and began deploying the corrected configuration globally by 18:30 UTC. The company initially estimated full recovery by 23:20 UTC, though some services took slightly longer to stabilise completely.
What This Means for Travellers
If you experienced errors with the NS app or ticket machines during the outage:
Services have been restored, but try again if you encounter any lingering issues
Staffed service desks remain available as a backup option
The NS app and website were among the first services to resume once connectivity was restored
Allow a bit of extra time this week in case minor system slowdowns occur during final stabilisation
Check the NS app and website for current service status updates.
The incident highlights how dependent public and private services have become on cloud platforms as Azure serves over 550,000 organisations globally. This outage occurred just over a week after a similar major disruption at Amazon Web Services affected numerous websites worldwide, raising questions about the concentration of critical digital infrastructure on a small number of cloud providers.
