From c34f56945ad591c8eae033ffe1038af038efa163 Mon Sep 17 00:00:00 2001 From: Mayne0213 Date: Sat, 10 Jan 2026 03:55:36 +0900 Subject: [PATCH] feat(prometheus): enable container CPU throttling metrics collection - Override default cAdvisorMetricRelabelings - Remove cfs_throttled_seconds_total from drop regex - Enables CPU Throttled panels in Grafana dashboards Co-Authored-By: Claude Opus 4.5 --- prometheus/helm-values.yaml | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+) diff --git a/prometheus/helm-values.yaml b/prometheus/helm-values.yaml index 144dedb..0f607e9 100644 --- a/prometheus/helm-values.yaml +++ b/prometheus/helm-values.yaml @@ -25,6 +25,28 @@ kubelet: replacement: "mayne-cluster" - sourceLabels: [__metrics_path__] targetLabel: metrics_path + # Override default drops to keep throttling metrics + cAdvisorMetricRelabelings: + # Drop unnecessary CPU metrics (but keep cfs_throttled_seconds_total) + - action: drop + regex: "container_cpu_(load_average_10s|system_seconds_total|user_seconds_total)" + sourceLabels: [__name__] + # Keep other default drops + - action: drop + regex: "container_fs_(io_current|io_time_seconds_total|io_time_weighted_seconds_total|reads_merged_total|sector_reads_total|sector_writes_total|writes_merged_total)" + sourceLabels: [__name__] + - action: drop + regex: "container_memory_(mapped_file|swap)" + sourceLabels: [__name__] + - action: drop + regex: "container_(file_descriptors|tasks_state|threads_max)" + sourceLabels: [__name__] + - action: drop + regex: "container_spec.*" + sourceLabels: [__name__] + - action: drop + regex: ".+;" + sourceLabels: [id, pod] # Resource metrics resourceRelabelings: - targetLabel: cluster