Empirical Root Cause Intelligence

Live Incident Post-Mortem & Outage Timeline Archive

Comprehensive technical post-mortems for major cloud, telecom, fintech, and gaming outages. Track empirical downtime durations, Mean Time to Recovery (MTTR), root cause analyses, and chronological mitigation milestones.

Documented Incidents
5 Verified Post-Mortems
Average MTTR Duration
1h 13m Mean Recovery
Critical Major Disruptions
3 Multi-Region P0s
Elevated Latency Events
2 P1 Partial Events
📡 GitHub 🟡 Elevated Degradation (P1) 📅 November 9, 2026 (11:42 - 13:06 UTC)

GitHub Actions Runner & Webhook Ingestion Degradation

MTTR Duration
1h 24m
🏷️ Queue Deadlock & Redis Worker Exhaustion 📍 US-East (N. Virginia), EU-West (Ireland)

GitHub experienced elevated webhook delivery delays and queue stagnation affecting GitHub Actions self-hosted and cloud runner dispatches across North America and Europe. Synthetic monitoring detected HTTP 504 gateway timeouts on api.github.com before recovery worker pools cleared the message backlog.

Timeline Milestones: 5 steps
Read Full Post-Mortem & Timeline →
📡 Cloudflare 🔴 Critical Outage (P0) 📅 November 5, 2026 (08:15 - 08:57 UTC)

Cloudflare 1.1.1.1 & CDN Edge Anycast BGP Anomaly

MTTR Duration
42m
🏷️ BGP Route Flapping & Transit Peer Leak 📍 Europe Central (Frankfurt, Amsterdam), US-East

Cloudflare experienced anomalous routing latency and intermittent packet loss across select European edge nodes due to upstream Tier-1 BGP route flapping. Automated traffic rerouting through secondary transit peers restored nominal latency profiles within 42 minutes.

Timeline Milestones: 4 steps
Read Full Post-Mortem & Timeline →
📡 Steam (Valve) 🟡 Elevated Degradation (P1) 📅 November 4, 2026 (23:00 - 23:55 UTC)

Steam Tuesday Maintenance: Extended Game Coordinator Stagnation

MTTR Duration
55m
🏷️ Game Coordinator Schema Migration Stagnation 📍 Global

Valve regular Tuesday maintenance window (normally 15–20 minutes) was extended to 55 minutes due to an unexpected schema lock during database migration for CS2 matchmaking coordinators. Store and friends list recovered promptly while match queues remained offline until coordinator locks cleared.

Timeline Milestones: 4 steps
Read Full Post-Mortem & Timeline →
📡 Chase Bank 🔴 Critical Outage (P0) 📅 October 28, 2026 (13:10 - 15:05 UTC)

Chase Mobile App Session Timeouts & Zelle Transfer Delays

MTTR Duration
1h 55m
🏷️ Database Deadlock & OAuth2 Token Service Outage 📍 United States (Nationwide)

Chase Bank customers across the United States experienced widespread mobile application login timeouts (Error 101) and delayed Zelle instant credit settlements (Error 202). Debit card point-of-sale authorizations and physical ATM withdrawals remained fully operational throughout.

Timeline Milestones: 4 steps
Read Full Post-Mortem & Timeline →
📡 Amazon Web Services (AWS) 🔴 Critical Outage (P0) 📅 October 14, 2026 (16:22 - 17:30 UTC)

AWS US-East-1 (N. Virginia) DynamoDB Latency & API Gateway 502s

MTTR Duration
1h 08m
🏷️ Internal DNS Subsystem Partition 📍 US-East-1 (Northern Virginia)

Multiple dependent enterprise platforms experienced cascading failures due to internal DNS partition in AWS us-east-1. DynamoDB write operations experienced elevated latency while API Gateway returned sporadic HTTP 502 Bad Gateway responses before internal route convergence.

Timeline Milestones: 4 steps
Read Full Post-Mortem & Timeline →

❓ Frequently Asked Questions: Incident Post-Mortems & MTTR

What is an Outage Post-Mortem and Root Cause Analysis (RCA)?

An outage post-mortem is a blameless technical dissection of an infrastructure disruption. It catalogs the chronological progression of an incident—from initial symptom detection and signal anomaly discovery, through engineering triage, root-cause identification (e.g. BGP route leaks, database deadlocks, DNS partitions), to final mitigation and systemic preventive action items.

How does IsDownNow calculate Mean Time to Recovery (MTTR)?

Mean Time to Recovery (MTTR) is calculated as the precise elapsed duration from the timestamp of the first sustained multi-vector diagnostic anomaly spike (elevated severity score >= 65) until synthetic probes and crowd report streams stabilize at nominal baseline (score <= 15).

How do these incidents impact provider Service Level Agreements (SLAs)?

Enterprise cloud and payment services typically guarantee 99.9% (Three Nines, max 43.8 minutes downtime per month) or 99.99% (Four Nines, max 4.38 minutes downtime per month). Incidents lasting over 45 minutes breach standard monthly SLA error budgets, qualifying enterprise customers for contract service credits.