On 20 October 2025 Amazon Web Services had a significant failure in us-east-1, its oldest and busiest region in Northern Virginia. The trigger was a DNS resolution problem affecting the DynamoDB endpoint, which then cascaded into the services that depend on it — including the internal systems AWS uses to launch new capacity. Recovery took most of a working day, with a long tail of backlogged queues after the underlying fault was fixed.
Why a US region matters to a Hampshire business
Most UK companies don't run their own servers in Virginia. They do, however, use dozens of SaaS products that quietly do. Authentication providers, chat tools, ticketing systems, payment processors, smart devices and mobile apps commonly keep at least one control-plane component in us-east-1. When it goes, your login page can fail even though your data lives in London.
What we saw on the helpdesk
- Sign-in loops on tools that use third-party identity providers
- Integrations silently failing and then replaying hours later
- File-sharing links returning errors while the app itself loaded
What to do about it
You can't engineer around another company's region, but you can reduce how much a bad day costs you:
- Know which of your tools are business-critical and check whether they publish a status page. Subscribe to it.
- Have an offline route for the two or three things that must keep working — usually phones, email and payments.
- Don't chain everything to one sign-in provider without a break-glass admin account you can use if it fails.
- Tell staff what to do during an outage so the first hour isn't spent rebooting laptops that were never the problem.
We monitor the vendors our customers depend on and post here when something widespread happens.