What happened
Amazon Web Services: Kinesis front-end fleet exceeded OS thread limit. Scope: Regional.
CHYP.ai · ORBIT // Cloudy_Day Security Suite
OUTAGE DISSECTOR · CASE FILE · Meltdown Index · vendor post-incident report
Amazon Web Services: Kinesis front-end fleet exceeded OS thread limit. Scope: Regional.
Severe storm
Drag or press Play to replay what was public at each moment. Each dot links to its evidence. Lanes are kept separate on purpose.
Amazon Web Services: Kinesis front-end fleet exceeded OS thread limit. Scope: Regional.
Amazon Kinesis, Amazon Cognito, Amazon CloudWatch, AWS Auto Scaling and 1 more.
136 catalogued services run on Amazon Web Services (inferred). Hit: Regional.
Resolved after 17h 8m.
Center: Amazon Web Services. Inner ring: what it runs on. Outer ring: catalogued services that run on it, coloured by their official status right now. Dependencies are inferred from public information.
Places come from names in the public text (cities, PoP codes, countries, regions). Approximate.
Failure modes as stated in the vendor's own post-incident report (Meltdown Index).
Correlation is not causation
Five independent lanes across the window. Brighter = stronger item. Items landing in the same column are coincidence unless the provider says otherwise.
The new capacity had caused all of the servers in the fleet to exceed the maximum number of threads allowed by an operating system configuration. As this limit was being exceeded, cache construction was failing to complete and front-end servers were ending up with useless shard-maps that left them unable to route requests to back-end clusters.
Duration is from first alarms at 5:15 AM PST to “Kinesis fully returned to normal at 10:23 PM PST,” per the PES. Capacity addition that triggered the event began at 2:44 AM PST.
Verbatim excerpt from Summary of the Amazon Kinesis Event in the Northern Virginia (US-EAST-1) Region.
100% similar · same provider
Lambda cell crossed a capacity threshold and hit a latent defect
CapacityControl planeCascade
85% similar · same provider
DynamoDB DNS automation wrote an empty record it could not repair
Control planeCascade
77% similar · same provider
Kinesis cell-management misidentified healthy hosts during a deployment
Config pushCapacityCascade
Nearest matches in the Control-Plane Meltdown Index (1.0.0, vendor-sourced, tracked since 2024 with earlier landmark cases). Similarity = provider + failure modes + scope.
Automated take: rules written by Cloudy_Day, applied to public data. Not a human verdict on this incident, not a root-cause finding, not affiliated with Amazon Web Services.
Everything above as one shareable, citable capsule with a content hash.
JSON capsuleMarkdownShare card (SVG)
Open Amazon Web Services in the War Room →Status page →
<a href="https://www.chyp.ai/dissect/m/aws-kinesis-2020-11-25/"><img src="https://www.chyp.ai/dissect/m/aws-kinesis-2020-11-25/card.svg" alt="Kinesis front-end fleet exceeded OS thread limit · ORBIT Outage Dissector" width="600"></a>