Writer: Jordan Sprogis
Editor: Lillian Castro
Reviewer: Cristian Lopez
Key Takeaways
- A Google Cloud IAM failure caused mass outages across Cloudflare, AWS, Azure, and dozens of popular services on June 12.
- Starting around 2 p.m. EST on June 12, Spotify, Discord, and Google services logged tens of thousands of outage reports, with full recovery taking 2 and a half hours in some regions.
Massive simultaneous outages occurred across Cloudflare, Google Cloud, AWS, and Azure on Thursday, June 12, disrupting tens of thousands of people’s access to dozens of popular apps and websites .
Authentication and access systems malfunctioned at about 1:51 p.m EST on June 12, which caused domino-effect failures across Google Cloud, Cloudflare, AWS, and Azure infrastructure.
Each platform had very similar patterns: a high rise of outages around 2:30 p.m., with a slow decline starting around 3:45 p.m., though many sites were down in several regions.
Here are the peak outage reports, according to Downdetector:
- Google Cloud logged more than 15,000 outage reports
- AWS: 6,000+
- Cloudflare: 3,000+
- Azure: 1,000+
For Google, the root cause of Thursday’s mass outage was a failure in its Identity and Access Management (IAM) service. It also hit major services like Spotify (46,000+ reports) and Discord (11,000+ reports).
As for why platforms like AWS, Azure, and Cloudflare were also affected, it isn’t yet known. There’s no public evidence that they rely on Google Cloud or vice versa.
Google Cloud’s most recent update was posted at 4:16 p.m.:
We have identified the root cause and applied appropriate mitigations. Our infrastructure has recovered in all regions except us-central1.
Google Cloud products that rely on the affected infrastructure are seeing recovery in multiple locations.
Our engineers are aware of the customers still experiencing issues on us-central1 and multi-region/us and are actively working on full recovery.
We do not have an ETA for full recovery.
We will provide an update by Thursday, 2025-06-12 14:00 PDT with current details.
Even after systems were fixed, it took about 2 and a half hours for full recovery.
It’s Not the First Time
Thursday’s outage is further fuels the mistrust in cloud providers. Although it’s widely used for its scalability, the cloud can sometimes act as a sitting duck.
Nearly half of all security breaches are cloud-based, and 96% of organizations admit they frequently face challenges with their cloud security strategies. Some notable past outages include:
- AWS Kinesis outage (2020): Broke dozens of services, including Adobe and Roku, for nearly a full day
- Azure global outage (2023): Took down Microsoft 365, Teams, and OneDrive for hours
- CrowdStrike meltdown (2024): Crashed 8.5 million Windows devices via faulty agent code
- Cloudflare attack (2024): A record-breaking 5.6 Tbps DDoS disrupted global service for hours
If providers can take anything from the June 12 outage, it’s that we’re looking at a weak ecosystem.
When a failure in a service like Google Cloud’s IAM can have a ripple effect on (seemingly) unrelated platforms, even those hosted elsewhere, it’s clear that today’s cloud architecture is more interconnected than most realize.
It’s a sign that cloud anxiety is real and that cross-cloud interdependence can be dangerous.
6/12/25 5:06 PM: This is a developing story. Some details may later be updated.
