Resources · Outage Watch

The next outage
starts with a routine change.

AI and infrastructure are changing faster than anyone can check. Teams, and the agents working beside them, need a live source of truth for what depends on what, so they know what could break before it does. Here is what recent outages were caused by, in the providers' own words.

6 of 8

outages in this log trace to a change, an update, maintenance, a resilience test or automation.

#1

Configuration and change management failures are the top cause of network-related outages (Uptime Institute, 2026).

57%

say their most recent major outage cost more than $100,000 (Uptime Institute, 2026).

The Log

Recent outages and what caused them.

Updated weekly. Last update Nov 18, 2026. Each cause links to the provider's statement or reporting on it.

Date Service What happened Stated cause Source
Nov 16, 2026 Chase online bankingMost of a day Logins, balances, transfers and Zelle failed for thousands of customers. Chase called it a technical issue, not a cyberattack. Not disclosed yet AP ↗
Nov 13, 2026 FAA air traffic control networkAbout 4 hours 45 minutes A router installed during routine work at a remote site flooded the FAA's telecom network; legacy equipment became unstable and departures stopped nationwide. Maintenance "a routine replacement at a remote facility" (FAA) FAA ↗
Nov 13, 2026 WorkdayAbout 18 hours A disaster-recovery test reached live databases and left data inconsistent, locking employees out of HR and payroll. Resilience testing "unexpectedly connected to Production databases" (Workday) UC Today ↗
Nov 12, 2026 Microsoft 365Nearly 24 hours Outlook, Teams and Copilot disrupted worldwide; a defect shipped in a July change surfaced months later. Software update "a latent defect" (Microsoft) ITPro ↗
Nov 10, 2026 Apple iCloud and Apple AccountAbout 2 hours Mail, Calendar, Photos and sign-in failed for users worldwide. Apple has not published a cause. Not disclosed yet StatusGator ↗
Nov 18, 2025 CloudflareAbout 6 hours Sites across the internet returned errors after an internal file doubled in size and crashed the proxy. Configuration change "a change to one of our database systems' permissions" (Cloudflare) Cloudflare ↗
Oct 29, 2025 Azure Front DoorAbout 8 hours Azure, Microsoft 365 and thousands of customer sites slowed or failed globally. Configuration change "an inadvertent configuration change" (Microsoft) ThousandEyes ↗
Oct 20, 2025 AWS us-east-1About 15 hours Two DNS automation processes collided and emptied a core DynamoDB endpoint, cascading across AWS. Automation "a latent race condition" (AWS) AWS ↗

Outage Watch

Know what could break. Before it does.

Free for one engineer, in the AI tool you already use. Share the map with your team on Pro.