Merge pull request prometheus-community#44 from frezes/feat/kube-prom…

…etheus-stack [kube-prometheus-stack] fix recording rule bugs
junotx · Feb 6, 2024 · 86d31a2 · 86d31a2
2 parents 41682fe + 180eaf9
commit 86d31a2
Show file tree

Hide file tree

Showing 4 changed files with 18 additions and 18 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: 52.1.10
+version: 52.1.11
 appVersion: v0.68.0
 kubeVersion: ">=1.19.0-0"
 home: https://github.com/prometheus-operator/kube-prometheus

diff --git a/...s/kube-prometheus-stack/templates/prometheus/rules-1.14/whizard-telemetry-node.rules.yaml b/...s/kube-prometheus-stack/templates/prometheus/rules-1.14/whizard-telemetry-node.rules.yaml
@@ -136,7 +136,7 @@ spec:
       {{- end }}
     - expr: |-
         sum by (cluster, node, device) (
-              max by (namespace, pod, instance, device) (
+              max by (cluster, namespace, pod, instance, device) (
                   node_filesystem_avail_bytes{device=~"/dev/.*",device!~"/dev/loop\\d+", job="node-exporter"}
               )
           * on (cluster, namespace, pod) group_left (node)
@@ -275,12 +275,12 @@ spec:
     - expr: |-
         count by (cluster, node) (
                   node_namespace_pod:kube_pod_info:{node!=""}
-                unless on (pod, namespace, cluster,)
+                unless on (pod, namespace, cluster)
                   (kube_pod_status_phase{job="kube-state-metrics",phase="Succeeded"} > 0)
-              unless on (pod, namespace, cluster,)
+              unless on (pod, namespace, cluster)
                 (
                     (kube_pod_status_ready{condition="true",job="kube-state-metrics"} > 0)
-                  and on (pod, namespace, cluster,)
+                  and on (pod, namespace, cluster)
                     (kube_pod_status_phase{job="kube-state-metrics",phase="Running"} > 0)
                 )
             unless on (cluster, pod, namespace)
@@ -289,7 +289,7 @@ spec:
         /
         count by (cluster, node) (
               node_namespace_pod:kube_pod_info:{node!=""}
-            unless on (pod, namespace, cluster,)
+            unless on (pod, namespace, cluster)
               kube_pod_status_phase{job="kube-state-metrics",phase="Succeeded"} > 0
         )
       record: node:pod_abnormal_utilisation:ratio

diff --git a/ks-prometheus/components/whizard-telemetry-mixin/rules/custom.libsonnet b/ks-prometheus/components/whizard-telemetry-mixin/rules/custom.libsonnet
@@ -144,7 +144,7 @@
             record: 'node:node_device_filesystem_available_bytes:sum',
             expr: |||
               sum by (%(clusterLabel)s, node, device) (
-                    max by (namespace, pod, instance, device) (
+                    max by (%(clusterLabel)s, namespace, %(podLabel)s, instance, device) (
                         node_filesystem_avail_bytes{%(hostFilesystemDeviceSelector)s, %(nodeExporterSelector)s}
                     )
                 * on (%(clusterLabel)s, namespace, %(podLabel)s) group_left (node)
@@ -170,7 +170,7 @@
             record: 'node:node_device_filesystem_bytes_total:sum',
             expr: |||
               sum by (%(clusterLabel)s, node, device) (
-                    max by (%(clusterLabel)s, namespace, pod, instance, device) (
+                    max by (%(clusterLabel)s, namespace, %(podLabel)s, instance, device) (
                         node_filesystem_size_bytes{%(hostFilesystemDeviceSelector)s, %(nodeExporterSelector)s}
                     )
                 * on (%(clusterLabel)s, namespace, %(podLabel)s) group_left (node)
@@ -229,21 +229,21 @@
             expr: |||
               count by (%(clusterLabel)s, node) (
                         node_namespace_pod:kube_pod_info:{node!=""}
-                      unless on (pod, namespace, %(clusterLabel)s,)
+                      unless on (%(podLabel)s, namespace, %(clusterLabel)s)
                         (kube_pod_status_phase{job="kube-state-metrics",phase="Succeeded"} > 0)
-                    unless on (pod, namespace, %(clusterLabel)s,)
+                    unless on (%(podLabel)s, namespace, %(clusterLabel)s)
                       (
                           (kube_pod_status_ready{condition="true",job="kube-state-metrics"} > 0)
-                        and on (pod, namespace, %(clusterLabel)s,)
+                        and on (%(podLabel)s, namespace, %(clusterLabel)s)
                           (kube_pod_status_phase{job="kube-state-metrics",phase="Running"} > 0)
                       )
-                  unless on (%(clusterLabel)s, pod, namespace)
+                  unless on (%(clusterLabel)s, %(podLabel)s, namespace)
                     kube_pod_container_status_waiting_reason{job="kube-state-metrics",reason="ContainerCreating"} > 0
               )
               /
               count by (%(clusterLabel)s, node) (
                     node_namespace_pod:kube_pod_info:{node!=""}
-                  unless on (pod, namespace, %(clusterLabel)s,)
+                  unless on (%(podLabel)s, namespace, %(clusterLabel)s)
                     kube_pod_status_phase{job="kube-state-metrics",phase="Succeeded"} > 0
               )
             ||| % $._config,

diff --git a/ks-prometheus/manifests/whizard-telemetry-prometheusRule.yaml b/ks-prometheus/manifests/whizard-telemetry-prometheusRule.yaml
@@ -106,7 +106,7 @@ spec:
       record: node:node_device_filesystem_utilisation:ratio
     - expr: |
         sum by (cluster, node, device) (
-              max by (namespace, pod, instance, device) (
+              max by (cluster, namespace, pod, instance, device) (
                   node_filesystem_avail_bytes{device=~"/dev/.*",device!~"/dev/loop\\d+", job="node-exporter"}
               )
           * on (cluster, namespace, pod) group_left (node)
@@ -161,12 +161,12 @@ spec:
     - expr: |
         count by (cluster, node) (
                   node_namespace_pod:kube_pod_info:{node!=""}
-                unless on (pod, namespace, cluster,)
+                unless on (pod, namespace, cluster)
                   (kube_pod_status_phase{job="kube-state-metrics",phase="Succeeded"} > 0)
-              unless on (pod, namespace, cluster,)
+              unless on (pod, namespace, cluster)
                 (
                     (kube_pod_status_ready{condition="true",job="kube-state-metrics"} > 0)
-                  and on (pod, namespace, cluster,)
+                  and on (pod, namespace, cluster)
                     (kube_pod_status_phase{job="kube-state-metrics",phase="Running"} > 0)
                 )
             unless on (cluster, pod, namespace)
@@ -175,7 +175,7 @@ spec:
         /
         count by (cluster, node) (
               node_namespace_pod:kube_pod_info:{node!=""}
-            unless on (pod, namespace, cluster,)
+            unless on (pod, namespace, cluster)
               kube_pod_status_phase{job="kube-state-metrics",phase="Succeeded"} > 0
         )
       record: node:pod_abnormal_utilisation:ratio