Coinbase suffers five‑hour outage

Coinbase users got a rough reminder of what “your funds are safe” does and does not mean. Safe is not the same as usable. Late on May 7, the exchange’s web and mobile apps slid into degraded performance, and for hours customers said they could not trade, move funds, or reliably access the platform. Coinbase tied the disruption to an AWS outage, then spent the night bringing markets back. (status.coinbase.com) ### What actually broke? The core problem was access to Coinbase itself. The company’s status page said customers could be “unable to transact on web and mobile” starting at 5:56 p.m. PDT on May 7. By 7:37 p.m., Coinbase said the issue was tied to an AWS outage. Later updates said all markets would first go into “Cancel Only” mode before trading was re-enabled, and the page finally showed service restored on May 8. (status.coinbase.com) ### How long was Coinbase impaired? Long enough to matter. If you use Coinbase during U.S. evening trading, this was basically a lost session. The incident started at 5:56 p.m. PDT on May 7, Coinbase was still posting updates at 9:52 p.m., and it did not begin re-enabling trading until 2:28 a.m. on May 8. That puts the disruption at well over five hours before markets started coming back. (status.coinbase.com) ### Why did AWS matter so much? Because this was not just a generic cloud hiccup. AWS said an affected availability zone in US-EAST-1 was dealing with increased temperatures, impaired EC2 and EBS resources, and knock-on errors and latency in dependent services. Reuters’ write-up framed it as a North Virginia data-center overheating event that disrupted some services. In (status.coinbase.com)s enough to freeze activity at one of the biggest U.S. crypto exchanges. (health.aws.amazon.com) ### Was this only a Coinbase problem? Mostly, from the user point of view. Plenty of companies can get nicked by an AWS incident, but Coinbase was the high-visibility consumer-facing casualty here because trading is so time-sensitive. If a shopping app stalls, you buy later. If an exchange stalls, users immediately start thinking about missed exits, missed entries, (health.aws.amazon.com)creenshots spread so fast. (status.coinbase.com) ### Why did traders react so hard? Because crypto never closes. There is no bell, no overnight pause, no clean reset. A five-hour outage on a 24/7 market feels less like a website bug and more like being locked out of your broker while prices keep moving. Coinbase kept repeating that funds were safe, but the real complaint was control — users could not act when they wanted to. (status.coinbase.com) ### Why is the timing awkward for Coinbase? The outage hit within hours of Coinbase reporting a weak quarter. The company posted a $1.49 per-share loss, versus expectations for a $0.27 profit, and revenue of $1.41 billion, below the $1.52 billion analysts expected. Shares fell in after-hours trading. So the company was already trying to explain softer trading conditions when its platform reliability became the story. (cnbc.com) ### What is the bigger lesson here? Centralized exchanges sell convenience, but the catch is dependency. You depend on the exchange, its app, its risk controls, and apparently its cloud architecture. AWS did not “lose” anyone’s coins here, but it exposed a different kind of fragility — concentration risk. If one infrastructure provider or one region can (cnbc.com)ole stack stays reachable under stress. (health.aws.amazon.com) ### Bottom line This outage was not a crypto price crash or a hack. It was more basic than that — a cloud infrastructure failure that turned access itself into the problem. And that is why it matters. For Coinbase, the hard question now is not whether funds were safe. It is whether users will keep trusting a platform that can still go dark when they most want to use it. (status.coinbase.com)