Infrastructure

Kubernetes · Network · Kafka · DB · Storage · 관측 파이프라인

관측 신호 분리

ztunnel(L4) · OpenTelemetry(L7) · Prometheus/KEDA(리소스) — 서로 다른 계층을 담당

ztunnel (L4)

서비스 연결량 · mTLS 상태 · 전송량 · L4 오류 — 네트워크 경로 자체의 건강도

OpenTelemetry (L7)

API latency · HTTP status · trace · AI 호출 구간 — 애플리케이션 동작

Prometheus / KEDA

CPU · Memory · Kafka lag · in-flight — 리소스와 확장 신호

mTLS 적용률

95.7%

1개 링크 미적용

분당 신규 연결

443conn/min

평균 L4 오류율

0.29%

안정

관측 대상 링크

23

서비스 메시 링크

ztunnel 기준 pod-to-pod L4 연결 (istio_tcp_* 지표)

연결mTLSOpened/minClosed/minSentReceivedL4 오류율
Marathon Service PostgreSQL미적용1615115.4KB/s242.0KB/s1.31%
Marathon Service RedisSTRICT272546.5KB/s177.6KB/s0.13%
Marathon Service marathon-events (Kafka)STRICT373511.4KB/s26.7KB/s2.39%
Marathon Service Notification ConsumerSTRICT15144.6KB/s12.3KB/s0.21%
Running Record Service PostgreSQLSTRICT171724.4KB/s59.6KB/s0.16%
Running Record Service RedisSTRICT131087.2KB/s147.7KB/s0.21%
Running Record Service running-record-events (Kafka)STRICT28275.2KB/s9.8KB/s0.1%
AI Assistant PostgreSQLSTRICT17165.1KB/s7.4KB/s0.3%
AI Assistant RedisSTRICT10824.8KB/s53.8KB/s0.04%
AI Assistant Course Recommendation AISTRICT1312566B/s1.4KB/s0.03%
Course Recommendation AI PostgreSQLSTRICT282613.2KB/s21.5KB/s0.28%
Course Recommendation AI RedisSTRICT11926.2KB/s34.3KB/s0.01%
Crew Service PostgreSQLSTRICT131117.8KB/s51.5KB/s0.02%
Crew Service RedisSTRICT212120.9KB/s98.4KB/s0.2%
Crew Service crew-events (Kafka)STRICT19184.4KB/s7.3KB/s0.19%
Crew Service Notification ConsumerSTRICT23231.9KB/s6.5KB/s0.23%
Notification Consumer RedisSTRICT27256.4KB/s26.2KB/s0.03%
Notification Consumer notification-events (Kafka)STRICT21202.1KB/s4.2KB/s0%
Community Feed PostgreSQLSTRICT131230.4KB/s14.7KB/s0.22%
Community Feed RedisSTRICT282549.2KB/s70.5KB/s0.16%
Payment Gateway PostgreSQLSTRICT17155.4KB/s16.7KB/s0.22%
Payment Gateway RedisSTRICT10814.6KB/s24.9KB/s0.18%
Payment Gateway payment-events (Kafka)STRICT19171.3KB/s1.5KB/s0.03%