[{"id":"IL-0188","title":{"en":"Rolling updates on EKS returned 502 errors for about 30 seconds because Kubernetes deleted old pods as soon as new pods passed readiness, before the ALB had marked the new pod targets healthy","ko":"EKS 롤링 업데이트 때 Kubernetes 는 새 파드가 Ready 가 되자마자 옛 파드를 지웠지만 ALB 는 새 파드를 아직 healthy 로 보지 않아, 배포마다 약 30초 동안 502 가 났다."},"org":"Yonghyun Kim","org_slug":"yonghyun-kim","date":"2025-11-09","severity":"unknown","root_cause":"deploy-rollout","url":"https://benchlet.com/incidentlib/app/cases/IL-0188","shared_components":["ingress"]},{"id":"IL-0086","title":{"en":"A deployment meant for a CircleCI test Kubernetes cluster also removed Traefik and Calico resources from the production cluster, taking down the CircleCI UI and runner.circleci.com","ko":"테스트 Kubernetes 클러스터용 배포가 운영 클러스터의 Traefik 과 Calico 리소스까지 지워 CircleCI UI 와 runner.circleci.com 이 1시간 54분 동안 내려갔다."},"org":"CircleCI","org_slug":"circleci","date":"2021-03-11","severity":"unknown","root_cause":"deploy-rollout","url":"https://benchlet.com/incidentlib/app/cases/IL-0086","shared_components":["ingress"]}]