From dbd9dcf0c624da6db94173143e7316e18391391a Mon Sep 17 00:00:00 2001 From: Dmitri Fedotov Date: Fri, 12 Sep 2025 15:22:55 +0300 Subject: [PATCH 1/2] [alerts] rm goroutines alerts --- .../alerts/kubernetes-kubelet.yaml | 26 ------------------- 1 file changed, 26 deletions(-) diff --git a/charts/kubernetes-operations/alerts/kubernetes-kubelet.yaml b/charts/kubernetes-operations/alerts/kubernetes-kubelet.yaml index 80ef172..50dfff2 100644 --- a/charts/kubernetes-operations/alerts/kubernetes-kubelet.yaml +++ b/charts/kubernetes-operations/alerts/kubernetes-kubelet.yaml @@ -77,32 +77,6 @@ groups: summary: Kubelet is full. {{- end }} -{{ if not (.Values.prometheusRules.disabled.KubeletHighNumberOfGoRoutines | default false) }} - - alert: KubeletHighNumberOfGoRoutines - expr: go_goroutines{job=~".*kubelet"} > {{ dig "KubeletHighNumberOfGoRoutines" "threshold" "5000" .Values.prometheusRules }} - for: {{ dig "KubeletHighNumberOfGoRoutines" "for" "5m" .Values.prometheusRules }} - labels: - severity: {{ dig "KubeletHighNumberOfGoRoutines" "severity" "warning" .Values.prometheusRules }} - playbook: https://github.com/cloudoperators/kubernetes-operations/playbooks/KubeletHighNumberOfGoRoutines.md - {{ include "kubernetes-operations.additionalRuleLabels" . | nindent 6 }} - annotations: - description: Kublet on `{{`{{ $labels.node }}`}}` might be unresponsive due to a high number of Go routines. - summary: High number of Go routines. -{{- end }} - -{{ if not (.Values.prometheusRules.disabled.KubeletHighNumberOfGoRoutinesPredicted | default false) }} - - alert: KubeletHighNumberOfGoRoutinesPredicted - expr: abs(predict_linear(go_goroutines{job=~".*kubelet"}[1h], 2*3600)) > {{ dig "KubeletHighNumberOfGoRoutines" "threshold" "10000" .Values.prometheusRules }} - for: {{ dig "KubeletHighNumberOfGoRoutinesPredicted" "for" "5m" .Values.prometheusRules }} - labels: - severity: {{ dig "KubeletHighNumberOfGoRoutinesPredicted" "severity" "warning" .Values.prometheusRules }} - playbook: https://github.com/cloudoperators/kubernetes-operations/playbooks/KubeletHighNumberOfGoRoutinesPredicted.md - {{ include "kubernetes-operations.additionalRuleLabels" . | nindent 6 }} - annotations: - description: Kublet on `{{`{{$labels.node}}`}}` might become unresponsive due to a high number of go routines within 2 hours. - summary: Predicting high number of Go routines. -{{- end }} - {{ if not (.Values.prometheusRules.disabled.KubeletManyRequestErrors | default false) }} - alert: KubeletManyRequestErrors expr: | From 9edb39d3177dce3629a28d5e65acbf79d7408d73 Mon Sep 17 00:00:00 2001 From: Dmitri Fedotov Date: Fri, 12 Sep 2025 15:24:22 +0300 Subject: [PATCH 2/2] bump chart version --- charts/kubernetes-operations/Chart.yaml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/charts/kubernetes-operations/Chart.yaml b/charts/kubernetes-operations/Chart.yaml index e72bc37..be97bf4 100644 --- a/charts/kubernetes-operations/Chart.yaml +++ b/charts/kubernetes-operations/Chart.yaml @@ -3,7 +3,7 @@ apiVersion: v2 name: kubernetes-operations -version: 1.2.5 +version: 1.2.6 description: A set of Plutono dashboards and Prometheus alerting rules combined with playbooks to ensure effective operations of Kubernetes. maintainers: - name: richardtief