About AWSDown

AWSDown answers one question faster and more clearly than anyone else: is AWS down right now? - and then explains what broke, why, and what to do about it. The site is written for the people who get paged: developers, SREs, and the CTOs who own the postmortem.

How status is measured

The status you see is generated by our own ingestion pipeline, which polls public AWS health data (the AWS Health Dashboard) every 60 seconds, normalizes events per service and region, and stores every incident with its full update timeline. The live pill on the homepage and the per-service pages on /status reflect the most recent check, with the timestamp shown next to the pill. We do not guess from social media reports.

“Operational” means our latest check found no active incident for that service. Severity levels (minor / major / critical) follow the impact language AWS itself uses, and each service page charts the last 90 days of incident history so a green pill has context.

How post-mortems are researched

Analyses in the outage archive are compiled from primary sources: AWS’s official post-event summaries, the incident’s own status-page timeline, and reproducible technical detail from engineering write-ups. Every post-mortem lists its sources, shows a published and last-updated date, and separates established fact from our analysis. When AWS publishes a correction, we update the analysis and bump the visible date.

Who is behind this

AWSDown is operated by the team at Next Signal, whose monitoring platform powers the status checks and the free outage alerts. That relationship is disclosed wherever Next Signal is recommended. AWSDown is not affiliated with, sponsored by, or endorsed by Amazon Web Services; AWS service names are used for identification.

Corrections

If we got a fact wrong, we want to know. Corrections are applied in place with the last-updated date bumped, so the record is transparent.