Merge pull request #1205 from zuzzas/cpu-rules

Fixed CPU accounting in recording rules

assets/prometheus/rules/node.rules.yaml

@@ -2,7 +2,7 @@ groups:
 - name: node.rules
   rules:
   - record: instance:node_cpu:rate:sum
-    expr: sum(rate(node_cpu{mode!="idle",mode!="iowait",mode!~"^(?:guest.*)$"}[3m]))
+    expr: sum(rate(node_cpu{mode!="idle",mode!="iowait"}[3m]))
       BY (instance)
   - record: instance:node_filesystem_usage:sum
     expr: sum((node_filesystem_size{mountpoint="/"} - node_filesystem_free{mountpoint="/"}))
@@ -12,10 +12,10 @@ groups:
   - record: instance:node_network_transmit_bytes:rate:sum
     expr: sum(rate(node_network_transmit_bytes[3m])) BY (instance)
   - record: instance:node_cpu:ratio
-    expr: sum(rate(node_cpu{mode!="idle"}[5m])) WITHOUT (cpu, mode) / ON(instance)
+    expr: sum(rate(node_cpu{mode!="idle",mode!="iowait"}[5m])) WITHOUT (cpu, mode) / ON(instance)
       GROUP_LEFT() count(sum(node_cpu) BY (instance, cpu)) BY (instance)
   - record: cluster:node_cpu:sum_rate5m
-    expr: sum(rate(node_cpu{mode!="idle"}[5m]))
+    expr: sum(rate(node_cpu{mode!="idle",mode!="iowait"}[5m]))
   - record: cluster:node_cpu:ratio
     expr: cluster:node_cpu:rate5m / count(sum(node_cpu) BY (instance, cpu))
   - alert: NodeExporterDown

manifests/prometheus/prometheus-k8s-rules.yaml

@@ -498,7 +498,7 @@ data:
     - name: node.rules
       rules:
       - record: instance:node_cpu:rate:sum
-        expr: sum(rate(node_cpu{mode!="idle",mode!="iowait",mode!~"^(?:guest.*)$"}[3m]))
+        expr: sum(rate(node_cpu{mode!="idle",mode!="iowait"}[3m]))
           BY (instance)
       - record: instance:node_filesystem_usage:sum
         expr: sum((node_filesystem_size{mountpoint="/"} - node_filesystem_free{mountpoint="/"}))
@@ -508,10 +508,10 @@ data:
       - record: instance:node_network_transmit_bytes:rate:sum
         expr: sum(rate(node_network_transmit_bytes[3m])) BY (instance)
       - record: instance:node_cpu:ratio
-        expr: sum(rate(node_cpu{mode!="idle"}[5m])) WITHOUT (cpu, mode) / ON(instance)
+        expr: sum(rate(node_cpu{mode!="idle",mode!="iowait"}[5m])) WITHOUT (cpu, mode) / ON(instance)
           GROUP_LEFT() count(sum(node_cpu) BY (instance, cpu)) BY (instance)
       - record: cluster:node_cpu:sum_rate5m
-        expr: sum(rate(node_cpu{mode!="idle"}[5m]))
+        expr: sum(rate(node_cpu{mode!="idle",mode!="iowait"}[5m]))
       - record: cluster:node_cpu:ratio
         expr: cluster:node_cpu:rate5m / count(sum(node_cpu) BY (instance, cpu))
       - alert: NodeExporterDown