Server storm: global outage brings down Google, Spotify, YouTube, Amazon and other services
On the evening of June 12, users around the world experienced a massive outage of leading Internet services, including Google, Google Cloud, YouTube, Spotify, Discord, Amazon Web Services, Snapchat, Twitch, and others. The main reason was a simultaneous outage of infrastructure providers Cloudflare and Google Cloud, which led to denials of authorization and access to services.
This is evidenced by Downdetector data.
According to the service, the peak of complaints fell in the early afternoon – more than 40,000 messages for Spotify, more than 13,000 for Google Cloud and 11,000 for Discord. The outage began at approximately 1:19 p.m. Central American time (7:19 p.m. Kyiv time) and covered a wide geography.
Cloudflare confirmed the issue on its status portal — the outages were caused by errors in Access and WARP authentication , which caused the network platforms to stop processing requests properly. Their technical department announced that they had launched an investigation.
In parallel, Google Cloud said that the outage occurred in the “Identity and Access Management” service, which made it difficult to access BigQuery, App Engine, Firebase and other tools. The problem was noticed at around 19:51 Kyiv time (15:51 ET); by 21:30 most regions had already partially recovered.
Affected services:
- Google Search, Maps, Gmail;
- YouTube;
- Spotify;
- Discord;
- AWS;
- Snapchat;
- Twitch;
- Character AI, OpenAI.
While Google is not entirely dependent on Cloudflare, it does use their security network and CDN, which explains the scale of the impact.
Google Cloud and Cloudflare network engineers are continuing to work to resolve the impact. A return to stability is expected in the coming hours or by morning in some regions.
Recall that on June 10, the popular chatbot ChatGPT from OpenAI experienced a large-scale outage that affected users around the world.
Also follow “Pryamim” on Facebook , Twitter , Telegram , and Instagram.