On July 14, 2025, many people around the world suddenly found that they couldn’t access websites or use the internet normally. The reason? A temporary issue with Cloudflare’s free DNS service, called 1.1.1.1.
The outage started at 21:52 UTC and lasted about an hour, ending at 22:54 UTC. During that time, most users who rely on 1.1.1.1 for faster and safer browsing were affected.
What Is 1.1.1.1?
Think of 1.1.1.1 as a phonebook for the internet. When you type a website like google.com, your device asks a DNS server like 1.1.1.1, “What’s the IP address for this website?” If the DNS server isn’t working, your device doesn’t know where to go—so websites just won’t load.
Cloudflare has offered the 1.1.1.1 service since 2018. It’s free, fast, and private, and millions of people around the world use it every day.
Why Did 1.1.1.1 Stop Working?
The problem was caused by an internal mistake, not a hack or cyberattack.
Cloudflare was updating part of their system to prepare for a new service. While doing this, they accidentally included 1.1.1.1’s address in a section of the system where it didn’t belong. Nothing broke at that time, so they didn’t notice anything wrong.
But on July 14, they made another update that triggered a global refresh of settings. That’s when things went wrong: 1.1.1.1 was suddenly treated as if it was part of a new, inactive service. As a result, many of Cloudflare’s servers around the world stopped offering the 1.1.1.1 DNS service.
What Was the Impact?
- Millions of users around the world couldn’t load websites because 1.1.1.1 wasn’t responding.
- The problem lasted about 1 hour.
- Some users were not affected if they used DNS over HTTPS (DoH) with
cloudflare-dns.com, since it uses a different system. - Cloudflare detected the issue, declared it an incident, and started fixing it right away.
How Did Cloudflare Fix It?
At 22:20 UTC, they began undoing the bad configuration. This fix helped most of the servers get back online quickly. But about 23% of their servers needed extra time to reset, which is part of their normal safety process to avoid causing other problems.
By 22:54 UTC, the DNS service was fully restored around the world.