added old recording rules to defaults

Author: wtripp180901

ansible/roles/kube_prometheus_stack/defaults/main/helm.yml

@@ -91,7 +91,7 @@ kube_prometheus_stack_release_defaults:
       nodeSelector:
         clusterrole: "server"
 
-  additionalPrometheusRulesMap: "{{ prometheus_alert_rules }}"
+  additionalPrometheusRulesMap: "{{ prometheus_rules }}"
 
   grafana:
     service:

ansible/roles/kube_prometheus_stack/defaults/main/main.yml

@@ -126,13 +126,13 @@ prometheus_scrape_configs:
 #   - prometheus/targets/*.yml
 #   - prometheus/targets/*.json
 
-prometheus_extra_alert_rules: []
+prometheus_extra_rules: []
 
-prometheus_alert_rules:
+prometheus_rules:
   appliance-rules:
     groups:
     - name: all
-      rules: "{{ prometheus_extra_alert_rules }}" 
+      rules: "{{ prometheus_extra_rules }}" 
 
 # ------------------------------------------------------------------------------------------
 

environments/common/inventory/group_vars/all/prometheus.yml

@@ -23,11 +23,25 @@ prometheus_scrape_configs_default:
     replacement:   '${1}'
 
 prometheus_scrape_configs: "{{ prometheus_scrape_configs_default + (openondemand_scrape_configs if groups['openondemand'] | count > 0 else []) }}"
-prometheus_extra_alert_rules:
+prometheus_extra_rules:
   - alert: SlurmNodeDown
     annotations:
       message: '{% raw %}{{ $value }} Slurm nodes are in down status.{% endraw %}'
       summary: 'At least one Slurm node is down.'
     expr: "slurm_nodes_down > 0\n"
     labels:
       severity: critical
+  - record: node_cpu_system_seconds:record
+    expr: (100 * sum by(instance)(increase(node_cpu_seconds_total{mode="system",job="node-exporter"}[60s]))) / (sum by(instance)(increase(node_cpu_seconds_total{job="node-exporter"}[60s])))
+  - record: node_cpu_user_seconds:record
+    expr: (100 * sum by(instance)(increase(node_cpu_seconds_total{mode="user",job="node-exporter"}[60s]))) / (sum by(instance)(increase(node_cpu_seconds_total{job="node-exporter"}[60s])))
+  - record: node_cpu_iowait_seconds:record
+    expr: (100 * sum by(instance)(increase(node_cpu_seconds_total{mode="iowait",job="node-exporter"}[60s]))) / (sum by(instance)(increase(node_cpu_seconds_total{job="node-exporter"}[60s])))
+  - record: node_cpu_other_seconds:record
+    expr: (100 * sum by(instance)(increase(node_cpu_seconds_total{mode!="idle",mode!="user",mode!="system",mode!="iowait",job="node-exporter"}[60s]))) / (sum by(instance)(increase(node_cpu_seconds_total{job="node-exporter"}[60s])))
+  - record: node_cpu_scaling_frequency_hertz_avg:record # frequency rules aren't working
+    expr: avg by (instance) (node_cpu_scaling_frequency_hertz)
+  - record: node_cpu_scaling_frequency_hertz_min:record
+    expr: min by (instance) (node_cpu_scaling_frequency_hertz)
+  - record: node_cpu_scaling_frequency_hertz_max:record
+    expr: max by (instance) (node_cpu_scaling_frequency_hertz)