On 2026-11-11 the desktop booted with the RTC read as UTC while it held KST, so the clock ran 9 hours ahead until NTP pulled it back. Prometheus had written one hour of samples stamped 18:36-19:36Z into its TSDB head during that window. With outOfOrderTimeWindow at the chart default of 0s, every later scrape was older than the head's valid range and was dropped as "out of bounds". From 11:21Z no metric was stored, no rule could fire on real data, and KubeAPIDown (an absent() rule) fired falsely. It would have stayed blind until the wall clock passed the bogus head time, about 7 hours later. Set prometheusSpec.tsdb.outOfOrderTimeWindow to 10h. Samples up to 10 hours behind the head are now accepted, which covers the +9h KST-as-UTC misread. The README records why. Rejected: deleting the head WAL from the Retain PV. It works at once but removes persisted data and needs the pod stopped. Rejected: waiting for the clock to catch up. It leaves monitoring blind for hours during the period when the cause of the incident is still being found. Risk: out-of-order samples use an extra in-memory head and vertical compaction only while they arrive; in normal operation none arrive. A genuinely late exporter could now backfill up to 10h, which is harmless on this single-node Prometheus. Preventing the clock jump itself is the host change in infrastructure/host, not this commit. Verification: the values file parses and holds prometheus.prometheusSpec.tsdb.outOfOrderTimeWindow=10h. The live Prometheus CR already carried tsdb.outOfOrderTimeWindow=0s from this chart key, so the key maps to spec.tsdb. The same value was patched onto the live CR: the operator wrote out_of_order_time_window: 10h, the config reloaded at 13:04:37Z, "too old" warnings dropped to 0, all 34 targets report up, and KubeAPIDown resolved. The pinned renderer (scripts/validate/render-observability-core.sh) was not run: it requires Helm v3.19.4 and kustomize v5.8.1, which are not installed on the laptop or the desktop.
kube-prometheus-stack
고정 chart 88.0.1을 observability namespace에 렌더한다. Grafana와
node-exporter, control-plane에 존재하지 않는 scheduler/controller-manager/etcd/
kube-proxy monitor는 끈다. Prometheus와 Alertmanager 데이터는 각각의 Retain
Local PV에 보존하며 CRD는 별도 root에서 먼저 적용한다.
Prometheus discovery는 namespace label
observability.hyeonworks.com/enabled=true와 resource label
observability.hyeonworks.com/instance=home의 교집합만 선택한다.
Alertmanager source는 selector discovery 대신 global
alertmanagerConfiguration.name=platform-alertmanager를 참조한다. 참조 대상의
Slack Secret이 준비되기 전에는 이 변경을 live에 적용하지 않고 기존 null 구성을
유지한다.
Prometheus TSDB는 outOfOrderTimeWindow: 10h로 늦게 도착한 샘플을 받는다. 시계가
앞으로 튀었다가 되돌아오면 TSDB 머리에 미래 시각 샘플이 남고, 창이 0이면 그 시각이
지날 때까지 모든 수집이 out of bounds로 버려진다. 메트릭과 경보가 함께 멈추고,
absent() 경보(KubeAPIDown 등)는 거짓으로 울린다. 10시간은 재부팅 때 RTC의 KST를
UTC로 읽는 +9시간 오차를 덮는 값이다(2026-11-11 사고). 늦은 샘플이 없으면 추가 메모리는
거의 들지 않는다. 시계 자체를 막는 설정은 infrastructure/host가 맡는다.