Amazon cloud outage takes down Netflix, Instagram, Pinterest, & more
An outage of Amazon's Elastic Compute Cloud in North Virginia has taken down Netflix, Pinterest, Instagram, and other services. According to numerous Twitter updates and our own checks, all three services are unavailable as of Friday evening at 9:10 p.m. PT.
Context & Ripple Effects
The timing sharpens the story: this outage lands days after [[a:1196542|TechCrunch asked whether Instagram and other sites could avoid going down with Amazon's ship]], and hours after a storm hit AWS's North Virginia hub — the region hosting all three affected services. The pickup was unusually broad for an infrastructure incident, with The Verge, GeekWire, GigaOM, Engadget, Forbes, and The Next Web all carrying it the same evening.
Each named casualty had just been investing visibly in its cloud footprint: Netflix open sourced its Asgard deployment tool on GitHub on June 25, while Pinterest — which overtook Tumblr in US web traffic on June 28 — was scaling fast enough that downtime carried real audience cost.
First-order effects
- Netflix, Instagram, and Pinterest users lost service Friday evening, with Twitter updates standing in for official status pages as the primary incident channel.
- Instagram's outage came at a moment of peak product momentum, days after its version 2.5 release added Explore and Facebook Likes integration.
Second-order effects
- Rival consumer services hosted elsewhere gained a temporary audience window, since switching costs during an outage favor whoever is still up.
- The incident puts pressure on engineering teams at AWS's biggest tenants to justify multi-region redundancy budgets that looked like overkill a week earlier.
Third-order effects
- If single-region dependence keeps producing internet-wide blackouts, resilience architecture stops being an engineering preference and becomes a procurement requirement for any consumer service betting on AWS.
- Amazon's scale means its operational failures now register as industry events rather than vendor hiccups, raising the reputational stakes of every EC2 incident.
The trend: Consumer web services concentrating on a single cloud region are turning one provider's weather and hardware failures into industry-wide outages, forcing resilience architecture up the agenda.