DongHyeonkaandClaude Opus 5
dba0c3975c
Nine database round trips per login multiply the injected delay to 1.9 seconds, and connection pool queueing multiplies it again under twenty concurrent requests. The readiness probe joins the same queue and times out, so the node leaves the load balancer and pushes its load onto the one still standing. Two injections missed first: the guest interface is enp1s0 rather than eth0, and a filter on it can never match a pod IP because flannel has already encapsulated the packet. The delay has to go on flannel.1, before encapsulation. The predicted rise in optimistic lock conflicts did not happen, because logins insert new rows rather than contending for one. That belongs to B-3. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
21 lines
943 B
Plaintext
21 lines
943 B
Plaintext
=== 오버레이 인터페이스 확인 ===
|
|
flannel.1 UNKNOWN a6:b2:62:04:c1:a4 <BROADCAST,MULTICAST,UP,LOWER_UP>
|
|
cni0 UP 5a:77:1a:e2:b0:a4 <BROADCAST,MULTICAST,UP,LOWER_UP>
|
|
|
|
=== flannel.1 에 주입 — 여기서는 파드 IP 가 보인다 (캡슐화 전) ===
|
|
qdisc prio 1: root refcnt 2 bands 3 priomap 1 2 2 2 1 2 0 0 1 1 1 1 1 1 1 1
|
|
Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
|
|
backlog 0b 0p requeues 0
|
|
qdisc netem 30: parent 1:3 limit 1000 delay 200ms
|
|
Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
|
|
backlog 0b 0p requeues 0
|
|
|
|
=== [검증] 필터에 패킷이 걸리는가 ===
|
|
qdisc netem 30: parent 1:3 limit 1000 delay 200ms
|
|
Sent 18388 bytes 150 pkt (dropped 0, overlimits 0 requeues 0)
|
|
backlog 0b 0p requeues 0
|
|
|
|
=== 두 노드 지연 비교 (기준선: k0=70ms k1=66ms) ===
|
|
keycloak-0 평균 41 ms 최대 57 ms
|
|
keycloak-1 평균 1872 ms 최대 1887 ms
|