SShortSingh.
Back to feed

AWS US-EAST-1 Outage Knocked 3,500 Companies Offline Across 60 Countries

0
·2 views

A DNS race condition in Amazon Web Services' US-EAST-1 region on 20 October 2025 triggered a 15-hour cascade that disrupted over 75 AWS services and affected more than 3,500 companies across 60 countries. The failure brought down major platforms including Snapchat and Fortnite, froze banking transactions, and even caused internet-dependent smart beds to malfunction with no manual override available. The outage exposed a critical structural vulnerability: US-EAST-1 functions as a global control plane for AWS, meaning organisations that had built infrastructure in other regions were still dependent on a single data centre in Northern Virginia. AWS, Microsoft Azure, and Google Cloud together control over 62 percent of the global cloud market, with an estimated 94 percent of enterprise services relying on at least one of the three. Industry data shows IT downtime now costs large enterprises an average of $23,750 per minute, with global losses across major companies estimated at $400 billion annually.

Read the full story at DEV Community

This is an AI-generated summary. ShortSingh links to the original source for the complete article.

Discussion (0)

Log in to join the discussion and vote.

Log in

Related stories

0
ProgrammingDEV Community ·

Team Builds Memory-Retaining AI Tutor in Four Days Using Cognee at Hackathon

A development team created Continuum, an AI tutoring system, during a hackathon using Cognee as its memory layer alongside FastAPI and a large language model. Unlike conventional AI tutors that reset between sessions, Continuum is designed to remember student mistakes, recognize learning patterns, and adapt its teaching strategies over time. The system is built around four core memory operations: remembering student interactions, recalling past context, improving teaching preferences, and forgetting resolved misconceptions. The team structured the four-day build so that each day's work depended on the previous, starting with verifying the memory pipeline before layering on curriculum and tutoring logic. The project aimed to demonstrate the smallest viable version of a personalized tutor that could convincingly show continuity across student sessions.

0
ProgrammingDEV Community ·

Developer Discovers Core Reasoning Layer Was Silently Dropped During AI System Rebuild

A developer auditing their AI project, Origin (OLT-1), found that the entire reasoning layer designed in version 1 had never been carried over to version 2. The missing component, which included a thalamus router and two-stage reasoner, was meant to compose fired concepts into logical outcomes rather than relying on flat pattern lookups. While the original code files still existed, a codebase search confirmed zero references to the reasoning classes anywhere in the v2 production system. The gap had gone unnoticed for months as other parts of the system, such as the encoder and dispatcher, continued to receive updates. Because v2 had grown from 327 concepts to over 8,800, the developer concluded the reasoning layer could not simply be reinserted and would need to be fully retrained from scratch under a new multi-phase plan called Stage D.

0
ProgrammingDEV Community ·

How a single phone call sparked a two-year infrastructure overhaul cutting file duplication

A software engineer working on a digital content distribution platform in mid-2000s Japan recounts how a partner company's delivery architecture required the same file to be copied into dozens of separate per-storefront server directories, creating significant redundancy. When the partner's lead engineer called to request slower file pushes due to server strain, the engineer instead proposed a structural fix: letting the viewer fetch content directly using a storefront parameter rather than relying on pre-placed copies. The partner agreed to explore the idea, and over roughly two years the two teams completed the overhaul in two quiet, incremental steps. The first step eliminated per-storefront file duplication by switching to a single canonical copy looked up via URL parameter, dramatically shrinking delivery batch sizes. The experience highlighted how a polite but firm counter-proposal during a routine support call can lead to lasting architectural improvements benefiting both parties.

0
ProgrammingDEV Community ·

WishList Ninja Syncs Shopify Wishlist Data to Email Platforms in Real Time

A new Shopify app called WishList Ninja addresses a common gap where wishlist data remains siloed within the app and never reaches a merchant's email marketing platform. The app writes live profile properties — including an active wishlist flag, item count, value, and top saved item — directly to Klaviyo, Mailchimp, or Omnisend on every save or removal. A key engineering challenge was ensuring the has_active_wishlist flag resets to false whenever a wishlist empties, covering edge cases like purchases, account merges, and collection-only saves. Five of the app's seven email flow triggers are derived from diffing Shopify product update webhooks against saved items, eliminating the need for scheduled catalog scans. The storefront component weighs under 12KB of vanilla JavaScript, and the app is now available on the Shopify App Store.