Amazon Web Services said most of its service operations were "succeeding normally" on Monday, after a major shutdown affected online platforms, from banks to social media giants.
The disruption affected global operations of AWS and a chunk of the global internet.
Seattle-based AWS initially said the issue was due to “increased error rates and latencies” for a number of services in that region of the US, warning “customers may be unable to create or update support cases”.
The company's engineers were "actively working on both mitigating the issue and fully understanding the root cause”, it said earlier.
In its latest update, AWS said the underlying issue had been "fully mitigated and most AWS service operations are succeeding normally now. Some requests may be throttled while we work towards full resolution."
What services were affected?
The problem started in Amazon's so-called US-East-1 Region, its Service Health dashboard said, and affected more than 80 of its own services, plus banks, social media platforms and online games that AWS hosts.
The problem affected many of Amazon's most popular and most-used services, including the main Amazon website, its Alexa digital assistant, Amazon Music and Prime Video.
Messaging platforms WhatsApp and Snapchat were also affected, according to DownDetector, as were UAE telecoms companies Du and e& and popular video games Fortnite and Roblox.
Other games affected include Clash of Clans, Clash Royale, Wordle and Pokemon Go. The PlayStation Network, the Microsoft Store and teleconferencing service Zoom were also impacted alongside major banks and trading platforms.
Users began reporting issues at about 11am UAE time, according to DownDetector. At 1.27pm UAE time, AWS reported that it was starting to see “significant signs of recovery”.
According to DownDetector, the following services were affected:
- Abu Dhabi Commercial Bank
- Amazon
- Amazon Alexa
- Amazon Web Services
- Bank of Scotland
- Blink
- BT
- Canva
- Coinbase
- Du
- e&
- EE
- Emirates
- Epic Games Store
- Fortnite
- Gmail
- Grok
- Halifax
- HM Revenue and Customs
- Life360
- Lloyds Bank
- Mega
- My Fitness Pal
- Perplexity
- Ring
- Roblox
- Signal
- Slack
- Snapchat
- Splunk
- Virgin Media
- Vodafone
- Xero
- Zoom
The National has contacted AWS for comment.
'Moderate' financial hit
The disruption is significant because AWS underpins a vast segment of the digital economy, as thousands of enterprises, consumer services, financial platforms, apps and government services rely on its computing, storage and networking infrastructure, said Rania Gule, a senior market analyst at XS.com.
AWS is one of the world's biggest web-hosting providers, which hosts more than 76 million websites, storing location information, data and contact details, says the tracker Built With.
Ms Gule told The National: "From an economic viewpoint, even a few hours of downtime translate into lost transactions, decreased user engagement, delayed deliveries, stalled payments and downstream fallout. Because the incident is cascading and global in scale, the risk is amplified.
"In the short term, the financial hit will be moderate [given that enterprises are somewhat prepared for short disruptions], but the reputational and structural risk is elevated – it underscores the fragility of over-concentration in one cloud provider."
The outage also reflects how interdependent the digital ecosystem has become, with a single point of failure, David Minarsch, chief executive of Switzerland-based distributed ledger technology company Valory, said.
It "affects millions [of applications] simultaneously" within minutes, so "systems must be designed to maintain continuity under stress", he said.
Companies whose websites are hosted by AWS have assured their users the situation was being monitored closely.
“We're aware many users are currently unable to access Coinbase due to an AWS outage. Our team is working on the issue and we'll provide updates here. All funds are safe,” Coinbase said in a post on X.
Parse.ly, a data analytics platform whose services are hosted by AWS, said the disruption has caused “degraded performance or temporary unavailability”.
However, “no data loss is expected [and] once AWS restores stability, affected services will automatically recover”, a Parse.ly representative told The National.
