[{"id":"IL-0179","title":{"en":"With DEBUG logging and synchronous log output, a high-traffic service filled the 64 KB stdout pipe, so request threads blocked in write() while holding database connections and HikariCP ran dry, failing requests in 30-second bursts","ko":"DEBUG 로그가 쏟아지자 동기 로깅을 하던 요청 스레드들이 가득 찬 stdout 파이프에서 멈춘 채 DB 커넥션을 쥐고 있어, HikariCP 커넥션 풀이 고갈되고 요청이 30초 주기로 한꺼번에 실패했다."},"org":"MUSINSA","org_slug":"musinsa","date":"2026-10-01","severity":"unknown","root_cause":"config-change","url":"https://benchlet.com/incidentlib/app/cases/IL-0179","shared_components":["telemetry-agent"]},{"id":"IL-0528","title":{"en":"Cloud Pub/Sub could not publish or subscribe in ten regions, and some ordered subscriptions stayed stuck for hours","ko":"Cloud Pub/Sub 가 여러 리전에서 게시와 구독을 모두 못 했고 순서 지정 구독 일부는 그날 밤까지 메시지를 받지 못했다"},"org":"Google Cloud","org_slug":"google-cloud","date":"2025-01-08","severity":"unknown","root_cause":"config-change","url":"https://benchlet.com/incidentlib/app/cases/IL-0528","shared_components":["message-broker"]},{"id":"IL-0005","title":{"en":"A new telemetry service overloaded the Kubernetes API servers of large clusters and broke DNS-based service discovery for all OpenAI services","ko":"새 텔레메트리 서비스가 모든 노드에서 무거운 쿠버네티스 API 요청을 보내 대형 클러스터의 컨트롤 플레인이 무너졌고, DNS 기반 서비스 디스커버리가 깨져 OpenAI 의 모든 서비스가 멈췄다"},"org":"OpenAI","org_slug":"openai","date":"2024-12-11","severity":"unknown","root_cause":"config-change","url":"https://benchlet.com/incidentlib/app/cases/IL-0005","shared_components":["telemetry-agent"]},{"id":"IL-0407","title":{"en":"Azure OpenAI returned 5xx errors in 14 of 28 regions after an automated cleanup deleted its GPU VMs and endpoints","ko":"정리 자동화가 Azure OpenAI 의 GPU VM 과 모델 엔드포인트를 지워 28개 지역 중 14곳에서 5xx 오류가 났다"},"org":"Microsoft Azure","org_slug":"microsoft-azure","date":"2024-07-13","severity":"unknown","root_cause":"config-change","url":"https://benchlet.com/incidentlib/app/cases/IL-0407","shared_components":["capacity-manager"]},{"id":"IL-0153","title":{"en":"DOKS worker nodes went NotReady at 06:00 UTC on three days because a systemd timer in the Debian image ran apt upgrade and killed it mid kernel upgrade, breaking node networking","ko":"DOKS 1.27.6-do.0 과 1.28.2-do.0 워커 노드에서 매일 아침 도는 apt 자동 업그레이드가 15분 타임아웃에 걸려 강제 종료되면서 노드 네트워킹이 망가지고, 노드가 사흘 연속 NotReady 가 됐다."},"org":"DigitalOcean","org_slug":"digitalocean","date":"2023-10-08","severity":"unknown","root_cause":"config-change","url":"https://benchlet.com/incidentlib/app/cases/IL-0153","shared_components":["linux-kernel"]},{"id":"IL-0109","title":{"en":"A retried node rollover on a DFDS EKS cluster set the Auto Scaling group back to 15 instances, and the default termination policy removed every healthy worker node, leaving the cluster with no nodes","ko":"로그 수집을 고치려던 EKS 노드 롤오버가 꼬이면서 Auto Scaling 그룹이 정상 워커 노드를 모두 종료해 클러스터에 노드가 하나도 남지 않았다."},"org":"DFDS","org_slug":"dfds","date":"2023-02-14","severity":"sev1","root_cause":"config-change","url":"https://benchlet.com/incidentlib/app/cases/IL-0109","shared_components":["telemetry-agent"]},{"id":"IL-0461","title":{"en":"Every rollout of Datadog's metrics query service on Kubernetes raised errors that the logs blamed on DNS","ko":"메트릭 질의 서비스를 롤아웃할 때마다 DNS 오류가 나며 질의가 실패했다"},"org":"Datadog","org_slug":"datadog","date":"2022-04-13","severity":"unknown","root_cause":"config-change","url":"https://benchlet.com/incidentlib/app/cases/IL-0461","shared_components":["linux-kernel"]}]