Incident Library?

Incident Library?

Contributing factor ?

Automation amplification 자동화 증폭 · `automation-amplified`

2 cases · 2024 to 2025

Root cause ?Configuration change 1 Dependency failure 1

Component ?GPU driver 1 Model request routing 1 Node operating system image 1 Node pool 1 TPU 1 XLA compiler 1

Symptom ?GPU unavailable 1 Node NotReady 1 Misrouted requests 1 Response quality degradation 1

Cases

IL-0022 Three overlapping routing and TPU bugs degraded model responses — Anthropic, 2025, sev3
Root cause Configuration change
IL-0021 GKE GPU nodes came up without a working NVIDIA driver — Google Cloud, 2024, sev2
Root cause Dependency failure

View on the graph