[{"id":"IL-0240","title":{"en":"Cloudflare API requests failed and the dashboard slowed for hours after the authentication database lost its replicas","ko":"인증 데이터베이스가 과부하에 걸려 Cloudflare API 와 대시보드가 6시간 33분 동안 실패하고 느려졌다"},"org":"Cloudflare","org_slug":"cloudflare","date":"2020-11-02","severity":"unknown","root_cause":"dns-network","url":"https://benchlet.com/incidentlib/app/cases/IL-0240","shared_components":["failover-manager","database-replication","database","network-link"]},{"id":"IL-0128","title":{"en":"Network interfaces that dropped for a few seconds made NCCL time out and stalled or slowed LLM training on a cluster of more than 10,000 GPUs","ko":"1만 개 넘는 GPU 로 학습하는 클러스터에서 네트워크 인터페이스가 몇 초씩 끊겼다 붙는 플래핑이 NCCL 타임아웃을 일으켜 학습이 멈추거나 느려졌다."},"org":"ByteDance","org_slug":"bytedance","date":"2024-02-23","severity":"unknown","root_cause":"dns-network","url":"https://benchlet.com/incidentlib/app/cases/IL-0128","shared_components":["network-link"]},{"id":"IL-0590","title":{"en":"F5 load balancers cached a path MTU of 296 after Rackspace reboots, and Mailgun connections failed intermittently","ko":"Rackspace ORD 재부팅 뒤 Mailgun 연결이 간헐적으로 실패하고 이벤트 유실과 이메일 중복 발송이 났다"},"org":"Mailgun","org_slug":"mailgun","date":"2014-09-27","severity":"unknown","root_cause":"dns-network","url":"https://benchlet.com/incidentlib/app/cases/IL-0590","shared_components":["database-replication"]}]