[kube-prometheus-stack] Configure additional aggregation labels + k8s…

….rules split (#3883) * [kube-prometheus-stack] Configure additional aggregation labels Signed-off-by: Jan-Otto Kröpke <[email protected]> * Re-run sync Signed-off-by: Jan-Otto Kröpke <[email protected]> * Re-run sync Signed-off-by: Jan-Otto Kröpke <[email protected]> * Add new values for rules Signed-off-by: Jan-Otto Kröpke <[email protected]> * Add new values for rules Signed-off-by: Jan-Otto Kröpke <[email protected]> * Add new values for rules Signed-off-by: Jan-Otto Kröpke <[email protected]> * [kube-prometheus-stack] Fix names Signed-off-by: Jan-Otto Kröpke <[email protected]> * bump version Signed-off-by: Jan-Otto Kröpke <[email protected]> --------- Signed-off-by: Jan-Otto Kröpke <[email protected]> Signed-off-by: Jan-Otto Kröpke <[email protected]> Signed-off-by: Jan-Otto Kröpke <[email protected]> Co-authored-by: Gabriel Martinez <[email protected]>
prometheus-community · Nov 22, 2023 · 4c10243 · 4c10243
1 parent 49e3c3a
commit 4c10243
Show file tree

Hide file tree

Showing 27 changed files with 700 additions and 501 deletions.
diff --git a/charts/kube-prometheus-stack/Chart.yaml b/charts/kube-prometheus-stack/Chart.yaml
@@ -21,7 +21,7 @@ name: kube-prometheus-stack
 sources:
   - https://github.com/prometheus-community/helm-charts
   - https://github.com/prometheus-operator/kube-prometheus
-version: 54.1.0
+version: 54.2.0
 appVersion: v0.69.1
 kubeVersion: ">=1.19.0-0"
 home: https://github.com/prometheus-operator/kube-prometheus

diff --git a/charts/kube-prometheus-stack/hack/sync_prometheus_rules.py b/charts/kube-prometheus-stack/hack/sync_prometheus_rules.py
@@ -94,7 +94,13 @@ def new_representer(dumper, data):
     'config-reloaders': ' .Values.defaultRules.rules.configReloaders',
     'etcd': ' .Values.kubeEtcd.enabled .Values.defaultRules.rules.etcd',
     'general.rules': ' .Values.defaultRules.rules.general',
-    'k8s.rules': ' .Values.defaultRules.rules.k8s',
+    'k8s.rules.container_cpu_usage_seconds_total': ' .Values.defaultRules.rules.k8sContainerCpuUsageSecondsTotal',
+    'k8s.rules.container_memory_cache': ' .Values.defaultRules.rules.k8sContainerMemoryCache',
+    'k8s.rules.container_memory_rss': ' .Values.defaultRules.rules.k8sContainerMemoryRss',
+    'k8s.rules.container_memory_swap': ' .Values.defaultRules.rules.k8sContainerMemorySwap',
+    'k8s.rules.container_memory_working_set_bytes': ' .Values.defaultRules.rules.k8sContainerMemoryWorkingSetBytes',
+    'k8s.rules.container_resource': ' .Values.defaultRules.rules.k8sContainerResource',
+    'k8s.rules.pod_owner': ' .Values.defaultRules.rules.k8sPodOwner',
     'kube-apiserver-availability.rules': ' .Values.kubeApiServer.enabled .Values.defaultRules.rules.kubeApiserverAvailability',
     'kube-apiserver-burnrate.rules': ' .Values.kubeApiServer.enabled .Values.defaultRules.rules.kubeApiserverBurnrate',
     'kube-apiserver-histogram.rules': ' .Values.kubeApiServer.enabled .Values.defaultRules.rules.kubeApiserverHistogram',
@@ -417,7 +423,7 @@ def write_group_to_file(group, url, destination, min_kubernetes, max_kubernetes)
     rules = add_rules_per_rule_conditions(rules, group)
     # initialize header
     lines = header % {
-        'name': group['name'],
+        'name': sanitize_name(group['name']),
         'url': url,
         'condition': condition_map.get(group['name'], ''),
         'init_line': init_line,
@@ -426,7 +432,11 @@ def write_group_to_file(group, url, destination, min_kubernetes, max_kubernetes)
     }
 
     # rules themselves
-    lines += rules
+    lines += re.sub(
+        r'\s(by|on) ?\(',
+        r' \1 ({{ range $.Values.defaultRules.additionalAggregationLabels }}{{ . }},{{ end }}',
+        rules
+    )
 
     # footer
     lines += '{{- end }}'
@@ -541,6 +551,10 @@ def main():
     print("Finished")
 
 
+def sanitize_name(name):
+    return re.sub('[_]', '-', name).lower()
+
+
 def jsonnet_import_callback(base, rel):
     if "github.com" in base:
         base = os.getcwd() + '/vendor/' + base[base.find('github.com'):]

diff --git a/charts/kube-prometheus-stack/templates/prometheus/_rules.tpl b/charts/kube-prometheus-stack/templates/prometheus/_rules.tpl
@@ -8,7 +8,13 @@ rules:
   - "config-reloaders"
   - "etcd"
   - "general.rules"
-  - "k8s.rules"
+  - "k8s.rules.container_cpu_usage_seconds_total"
+  - "k8s.rules.container_memory_cache"
+  - "k8s.rules.container_memory_rss"
+  - "k8s.rules.container_memory_swap"
+  - "k8s.rules.container_memory_working_set_bytes"
+  - "k8s.rules.container_resource"
+  - "k8s.rules.pod_owner"
   - "kube-apiserver-availability.rules"
   - "kube-apiserver-burnrate.rules"
   - "kube-apiserver-histogram.rules"

diff --git a/charts/kube-prometheus-stack/templates/prometheus/rules-1.14/alertmanager.rules.yaml b/charts/kube-prometheus-stack/templates/prometheus/rules-1.14/alertmanager.rules.yaml
@@ -73,8 +73,8 @@ spec:
         # Without max_over_time, failed scrapes could create false negatives, see
         # https://www.robustperception.io/alerting-on-gauges-in-prometheus-2-0 for details.
           max_over_time(alertmanager_cluster_members{job="{{ $alertmanagerJob }}",namespace="{{ $namespace }}"}[5m])
-        < on (namespace,service,cluster) group_left
-          count by (namespace,service,cluster) (max_over_time(alertmanager_cluster_members{job="{{ $alertmanagerJob }}",namespace="{{ $namespace }}"}[5m]))
+        < on ({{ range $.Values.defaultRules.additionalAggregationLabels }}{{ . }},{{ end }}namespace,service,cluster) group_left
+          count by ({{ range $.Values.defaultRules.additionalAggregationLabels }}{{ . }},{{ end }}namespace,service,cluster) (max_over_time(alertmanager_cluster_members{job="{{ $alertmanagerJob }}",namespace="{{ $namespace }}"}[5m]))
       for: 15m
       {{- with .Values.defaultRules.keepFiringFor }}
       keep_firing_for: "{{ . }}"
@@ -137,7 +137,7 @@ spec:
         runbook_url: {{ .Values.defaultRules.runbookUrl }}/alertmanager/alertmanagerclusterfailedtosendalerts
         summary: All Alertmanager instances in a cluster failed to send notifications to a critical integration.
       expr: |-
-        min by (namespace,service, integration) (
+        min by ({{ range $.Values.defaultRules.additionalAggregationLabels }}{{ . }},{{ end }}namespace,service, integration) (
           rate(alertmanager_notifications_failed_total{job="{{ $alertmanagerJob }}",namespace="{{ $namespace }}", integration=~`.*`}[5m])
         /
           ignoring (reason) group_left rate(alertmanager_notifications_total{job="{{ $alertmanagerJob }}",namespace="{{ $namespace }}", integration=~`.*`}[5m])
@@ -171,7 +171,7 @@ spec:
         runbook_url: {{ .Values.defaultRules.runbookUrl }}/alertmanager/alertmanagerclusterfailedtosendalerts
         summary: All Alertmanager instances in a cluster failed to send notifications to a non-critical integration.
       expr: |-
-        min by (namespace,service, integration) (
+        min by ({{ range $.Values.defaultRules.additionalAggregationLabels }}{{ . }},{{ end }}namespace,service, integration) (
           rate(alertmanager_notifications_failed_total{job="{{ $alertmanagerJob }}",namespace="{{ $namespace }}", integration!~`.*`}[5m])
         /
           ignoring (reason) group_left rate(alertmanager_notifications_total{job="{{ $alertmanagerJob }}",namespace="{{ $namespace }}", integration!~`.*`}[5m])
@@ -205,8 +205,8 @@ spec:
         runbook_url: {{ .Values.defaultRules.runbookUrl }}/alertmanager/alertmanagerconfiginconsistent
         summary: Alertmanager instances within the same cluster have different configurations.
       expr: |-
-        count by (namespace,service,cluster) (
-          count_values by (namespace,service,cluster) ("config_hash", alertmanager_config_hash{job="{{ $alertmanagerJob }}",namespace="{{ $namespace }}"})
+        count by ({{ range $.Values.defaultRules.additionalAggregationLabels }}{{ . }},{{ end }}namespace,service,cluster) (
+          count_values by ({{ range $.Values.defaultRules.additionalAggregationLabels }}{{ . }},{{ end }}namespace,service,cluster) ("config_hash", alertmanager_config_hash{job="{{ $alertmanagerJob }}",namespace="{{ $namespace }}"})
         )
         != 1
       for: 20m
@@ -238,11 +238,11 @@ spec:
         summary: Half or more of the Alertmanager instances within the same cluster are down.
       expr: |-
         (
-          count by (namespace,service,cluster) (
+          count by ({{ range $.Values.defaultRules.additionalAggregationLabels }}{{ . }},{{ end }}namespace,service,cluster) (
             avg_over_time(up{job="{{ $alertmanagerJob }}",namespace="{{ $namespace }}"}[5m]) < 0.5
           )
         /
-          count by (namespace,service,cluster) (
+          count by ({{ range $.Values.defaultRules.additionalAggregationLabels }}{{ . }},{{ end }}namespace,service,cluster) (
             up{job="{{ $alertmanagerJob }}",namespace="{{ $namespace }}"}
           )
         )
@@ -276,11 +276,11 @@ spec:
         summary: Half or more of the Alertmanager instances within the same cluster are crashlooping.
       expr: |-
         (
-          count by (namespace,service,cluster) (
+          count by ({{ range $.Values.defaultRules.additionalAggregationLabels }}{{ . }},{{ end }}namespace,service,cluster) (
             changes(process_start_time_seconds{job="{{ $alertmanagerJob }}",namespace="{{ $namespace }}"}[10m]) > 4
           )
         /
-          count by (namespace,service,cluster) (
+          count by ({{ range $.Values.defaultRules.additionalAggregationLabels }}{{ . }},{{ end }}namespace,service,cluster) (
             up{job="{{ $alertmanagerJob }}",namespace="{{ $namespace }}"}
           )
         )

diff --git a/charts/kube-prometheus-stack/templates/prometheus/rules-1.14/general.rules.yaml b/charts/kube-prometheus-stack/templates/prometheus/rules-1.14/general.rules.yaml
@@ -110,7 +110,7 @@ spec:
           '
         runbook_url: {{ .Values.defaultRules.runbookUrl }}/general/infoinhibitor
         summary: Info-level alert inhibition.
-      expr: ALERTS{severity = "info"} == 1 unless on(namespace) ALERTS{alertname != "InfoInhibitor", severity =~ "warning|critical", alertstate="firing"} == 1
+      expr: ALERTS{severity = "info"} == 1 unless on ({{ range $.Values.defaultRules.additionalAggregationLabels }}{{ . }},{{ end }}namespace) ALERTS{alertname != "InfoInhibitor", severity =~ "warning|critical", alertstate="firing"} == 1
       labels:
         severity: none
       {{- if or .Values.defaultRules.additionalRuleLabels .Values.defaultRules.additionalRuleGroupLabels.general }}

diff --git a/...us-stack/templates/prometheus/rules-1.14/k8s.rules.container_cpu_usage_seconds_total.yaml b/...us-stack/templates/prometheus/rules-1.14/k8s.rules.container_cpu_usage_seconds_total.yaml
@@ -0,0 +1,43 @@
+{{- /*
+Generated from 'k8s.rules.container-cpu-usage-seconds-total' group from https://raw.githubusercontent.com/prometheus-operator/kube-prometheus/main/manifests/kubernetesControlPlane-prometheusRule.yaml
+Do not change in-place! In order to change this file first read following link:
+https://github.com/prometheus-community/helm-charts/tree/main/charts/kube-prometheus-stack/hack
+*/ -}}
+{{- $kubeTargetVersion := default .Capabilities.KubeVersion.GitVersion .Values.kubeTargetVersionOverride }}
+{{- if and (semverCompare ">=1.14.0-0" $kubeTargetVersion) (semverCompare "<9.9.9-9" $kubeTargetVersion) .Values.defaultRules.create .Values.defaultRules.rules.k8sContainerCpuUsageSecondsTotal }}
+apiVersion: monitoring.coreos.com/v1
+kind: PrometheusRule
+metadata:
+  name: {{ printf "%s-%s" (include "kube-prometheus-stack.fullname" .) "k8s.rules.container-cpu-usage-seconds-total" | trunc 63 | trimSuffix "-" }}
+  namespace: {{ template "kube-prometheus-stack.namespace" . }}
+  labels:
+    app: {{ template "kube-prometheus-stack.name" . }}
+{{ include "kube-prometheus-stack.labels" . | indent 4 }}
+{{- if .Values.defaultRules.labels }}
+{{ toYaml .Values.defaultRules.labels | indent 4 }}
+{{- end }}
+{{- if .Values.defaultRules.annotations }}
+  annotations:
+{{ toYaml .Values.defaultRules.annotations | indent 4 }}
+{{- end }}
+spec:
+  groups:
+  - name: k8s.rules.container_cpu_usage_seconds_total
+    rules:
+    - expr: |-
+        sum by ({{ range $.Values.defaultRules.additionalAggregationLabels }}{{ . }},{{ end }}cluster, namespace, pod, container) (
+          irate(container_cpu_usage_seconds_total{job="kubelet", metrics_path="/metrics/cadvisor", image!=""}[5m])
+        ) * on ({{ range $.Values.defaultRules.additionalAggregationLabels }}{{ . }},{{ end }}cluster, namespace, pod) group_left(node) topk by ({{ range $.Values.defaultRules.additionalAggregationLabels }}{{ . }},{{ end }}cluster, namespace, pod) (
+          1, max by ({{ range $.Values.defaultRules.additionalAggregationLabels }}{{ . }},{{ end }}cluster, namespace, pod, node) (kube_pod_info{node!=""})
+        )
+      record: node_namespace_pod_container:container_cpu_usage_seconds_total:sum_irate
+      {{- if or .Values.defaultRules.additionalRuleLabels .Values.defaultRules.additionalRuleGroupLabels.k8sContainerCpuUsageSecondsTotal }}
+      labels:
+        {{- with .Values.defaultRules.additionalRuleLabels }}
+          {{- toYaml . | nindent 8 }}
+        {{- end }}
+        {{- with .Values.defaultRules.additionalRuleGroupLabels.k8sContainerCpuUsageSecondsTotal }}
+          {{- toYaml . | nindent 8 }}
+        {{- end }}
+      {{- end }}
+{{- end }}
diff --git a/...be-prometheus-stack/templates/prometheus/rules-1.14/k8s.rules.container_memory_cache.yaml b/...be-prometheus-stack/templates/prometheus/rules-1.14/k8s.rules.container_memory_cache.yaml
@@ -0,0 +1,42 @@
+{{- /*
+Generated from 'k8s.rules.container-memory-cache' group from https://raw.githubusercontent.com/prometheus-operator/kube-prometheus/main/manifests/kubernetesControlPlane-prometheusRule.yaml
+Do not change in-place! In order to change this file first read following link:
+https://github.com/prometheus-community/helm-charts/tree/main/charts/kube-prometheus-stack/hack
+*/ -}}
+{{- $kubeTargetVersion := default .Capabilities.KubeVersion.GitVersion .Values.kubeTargetVersionOverride }}
+{{- if and (semverCompare ">=1.14.0-0" $kubeTargetVersion) (semverCompare "<9.9.9-9" $kubeTargetVersion) .Values.defaultRules.create .Values.defaultRules.rules.k8sContainerMemoryCache }}
+apiVersion: monitoring.coreos.com/v1
+kind: PrometheusRule
+metadata:
+  name: {{ printf "%s-%s" (include "kube-prometheus-stack.fullname" .) "k8s.rules.container-memory-cache" | trunc 63 | trimSuffix "-" }}
+  namespace: {{ template "kube-prometheus-stack.namespace" . }}
+  labels:
+    app: {{ template "kube-prometheus-stack.name" . }}
+{{ include "kube-prometheus-stack.labels" . | indent 4 }}
+{{- if .Values.defaultRules.labels }}
+{{ toYaml .Values.defaultRules.labels | indent 4 }}
+{{- end }}
+{{- if .Values.defaultRules.annotations }}
+  annotations:
+{{ toYaml .Values.defaultRules.annotations | indent 4 }}
+{{- end }}
+spec:
+  groups:
+  - name: k8s.rules.container_memory_cache
+    rules:
+    - expr: |-
+        container_memory_cache{job="kubelet", metrics_path="/metrics/cadvisor", image!=""}
+        * on ({{ range $.Values.defaultRules.additionalAggregationLabels }}{{ . }},{{ end }}cluster, namespace, pod) group_left(node) topk by ({{ range $.Values.defaultRules.additionalAggregationLabels }}{{ . }},{{ end }}cluster, namespace, pod) (1,
+          max by ({{ range $.Values.defaultRules.additionalAggregationLabels }}{{ . }},{{ end }}cluster, namespace, pod, node) (kube_pod_info{node!=""})
+        )
+      record: node_namespace_pod_container:container_memory_cache
+      {{- if or .Values.defaultRules.additionalRuleLabels .Values.defaultRules.additionalRuleGroupLabels.k8sContainerMemoryCache }}
+      labels:
+        {{- with .Values.defaultRules.additionalRuleLabels }}
+          {{- toYaml . | nindent 8 }}
+        {{- end }}
+        {{- with .Values.defaultRules.additionalRuleGroupLabels.k8sContainerMemoryCache }}
+          {{- toYaml . | nindent 8 }}
+        {{- end }}
+      {{- end }}
+{{- end }}
diff --git a/...kube-prometheus-stack/templates/prometheus/rules-1.14/k8s.rules.container_memory_rss.yaml b/...kube-prometheus-stack/templates/prometheus/rules-1.14/k8s.rules.container_memory_rss.yaml
@@ -0,0 +1,42 @@
+{{- /*
+Generated from 'k8s.rules.container-memory-rss' group from https://raw.githubusercontent.com/prometheus-operator/kube-prometheus/main/manifests/kubernetesControlPlane-prometheusRule.yaml
+Do not change in-place! In order to change this file first read following link:
+https://github.com/prometheus-community/helm-charts/tree/main/charts/kube-prometheus-stack/hack
+*/ -}}
+{{- $kubeTargetVersion := default .Capabilities.KubeVersion.GitVersion .Values.kubeTargetVersionOverride }}
+{{- if and (semverCompare ">=1.14.0-0" $kubeTargetVersion) (semverCompare "<9.9.9-9" $kubeTargetVersion) .Values.defaultRules.create .Values.defaultRules.rules.k8sContainerMemoryRss }}
+apiVersion: monitoring.coreos.com/v1
+kind: PrometheusRule
+metadata:
+  name: {{ printf "%s-%s" (include "kube-prometheus-stack.fullname" .) "k8s.rules.container-memory-rss" | trunc 63 | trimSuffix "-" }}
+  namespace: {{ template "kube-prometheus-stack.namespace" . }}
+  labels:
+    app: {{ template "kube-prometheus-stack.name" . }}
+{{ include "kube-prometheus-stack.labels" . | indent 4 }}
+{{- if .Values.defaultRules.labels }}
+{{ toYaml .Values.defaultRules.labels | indent 4 }}
+{{- end }}
+{{- if .Values.defaultRules.annotations }}
+  annotations:
+{{ toYaml .Values.defaultRules.annotations | indent 4 }}
+{{- end }}
+spec:
+  groups:
+  - name: k8s.rules.container_memory_rss
+    rules:
+    - expr: |-
+        container_memory_rss{job="kubelet", metrics_path="/metrics/cadvisor", image!=""}
+        * on ({{ range $.Values.defaultRules.additionalAggregationLabels }}{{ . }},{{ end }}cluster, namespace, pod) group_left(node) topk by ({{ range $.Values.defaultRules.additionalAggregationLabels }}{{ . }},{{ end }}cluster, namespace, pod) (1,
+          max by ({{ range $.Values.defaultRules.additionalAggregationLabels }}{{ . }},{{ end }}cluster, namespace, pod, node) (kube_pod_info{node!=""})
+        )
+      record: node_namespace_pod_container:container_memory_rss
+      {{- if or .Values.defaultRules.additionalRuleLabels .Values.defaultRules.additionalRuleGroupLabels.k8sContainerMemoryRss }}
+      labels:
+        {{- with .Values.defaultRules.additionalRuleLabels }}
+          {{- toYaml . | nindent 8 }}
+        {{- end }}
+        {{- with .Values.defaultRules.additionalRuleGroupLabels.k8sContainerMemoryRss }}
+          {{- toYaml . | nindent 8 }}
+        {{- end }}
+      {{- end }}
+{{- end }}
diff --git a/...ube-prometheus-stack/templates/prometheus/rules-1.14/k8s.rules.container_memory_swap.yaml b/...ube-prometheus-stack/templates/prometheus/rules-1.14/k8s.rules.container_memory_swap.yaml
@@ -0,0 +1,42 @@
+{{- /*
+Generated from 'k8s.rules.container-memory-swap' group from https://raw.githubusercontent.com/prometheus-operator/kube-prometheus/main/manifests/kubernetesControlPlane-prometheusRule.yaml
+Do not change in-place! In order to change this file first read following link:
+https://github.com/prometheus-community/helm-charts/tree/main/charts/kube-prometheus-stack/hack
+*/ -}}
+{{- $kubeTargetVersion := default .Capabilities.KubeVersion.GitVersion .Values.kubeTargetVersionOverride }}
+{{- if and (semverCompare ">=1.14.0-0" $kubeTargetVersion) (semverCompare "<9.9.9-9" $kubeTargetVersion) .Values.defaultRules.create .Values.defaultRules.rules.k8sContainerMemorySwap }}
+apiVersion: monitoring.coreos.com/v1
+kind: PrometheusRule
+metadata:
+  name: {{ printf "%s-%s" (include "kube-prometheus-stack.fullname" .) "k8s.rules.container-memory-swap" | trunc 63 | trimSuffix "-" }}
+  namespace: {{ template "kube-prometheus-stack.namespace" . }}
+  labels:
+    app: {{ template "kube-prometheus-stack.name" . }}
+{{ include "kube-prometheus-stack.labels" . | indent 4 }}
+{{- if .Values.defaultRules.labels }}
+{{ toYaml .Values.defaultRules.labels | indent 4 }}
+{{- end }}
+{{- if .Values.defaultRules.annotations }}
+  annotations:
+{{ toYaml .Values.defaultRules.annotations | indent 4 }}
+{{- end }}
+spec:
+  groups:
+  - name: k8s.rules.container_memory_swap
+    rules:
+    - expr: |-
+        container_memory_swap{job="kubelet", metrics_path="/metrics/cadvisor", image!=""}
+        * on ({{ range $.Values.defaultRules.additionalAggregationLabels }}{{ . }},{{ end }}cluster, namespace, pod) group_left(node) topk by ({{ range $.Values.defaultRules.additionalAggregationLabels }}{{ . }},{{ end }}cluster, namespace, pod) (1,
+          max by ({{ range $.Values.defaultRules.additionalAggregationLabels }}{{ . }},{{ end }}cluster, namespace, pod, node) (kube_pod_info{node!=""})
+        )
+      record: node_namespace_pod_container:container_memory_swap
+      {{- if or .Values.defaultRules.additionalRuleLabels .Values.defaultRules.additionalRuleGroupLabels.k8sContainerMemorySwap }}
+      labels:
+        {{- with .Values.defaultRules.additionalRuleLabels }}
+          {{- toYaml . | nindent 8 }}
+        {{- end }}
+        {{- with .Values.defaultRules.additionalRuleGroupLabels.k8sContainerMemorySwap }}
+          {{- toYaml . | nindent 8 }}
+        {{- end }}
+      {{- end }}
+{{- end }}