Merge pull request #108 from frezes/feat/wiztelemetry-monitoring-helper

[wiztelemetry-monitoring-helper] fix recoridng rule

Author: junotx

charts/wiztelemetry-monitoring-helper/Chart.yaml

@@ -15,7 +15,7 @@ type: application
 # This is the chart version. This version number should be incremented each time you make changes
 # to the chart and its templates, including the app version.
 # Versions are expected to follow Semantic Versioning (https://semver.org/)
-version: 0.10.2
+version: 0.10.3
 
 # This is the version number of the application being deployed. This version number should be
 # incremented each time you make changes to the application. Versions are not expected to

charts/wiztelemetry-monitoring-helper/templates/wiztelemetry-rules/wiztelemetry-cluster.rules.yaml

@@ -26,14 +26,18 @@ spec:
     rules:
     - expr: |-
         max by ({{ range $.Values.defaultRules.additionalAggregationLabels }}{{ . }},{{ end }}cluster, node, ip, workspace, namespace, pod, qos_class, phase, workload, workload_type) (
-                  kube_pod_info{job="kube-state-metrics"}
-                * on ({{ range $.Values.defaultRules.additionalAggregationLabels }}{{ . }},{{ end }}cluster, namespace, pod) group_left (qos_class)
-                  max by ({{ range $.Values.defaultRules.additionalAggregationLabels }}{{ . }},{{ end }}cluster, namespace, pod, qos_class) (
-                    kube_pod_status_qos_class{job="kube-state-metrics"} > 0
-                  )
+                    kube_pod_info{job="kube-state-metrics"}
+                  * on ({{ range $.Values.defaultRules.additionalAggregationLabels }}{{ . }},{{ end }}cluster, namespace, pod) group_left (qos_class)
+                    max by ({{ range $.Values.defaultRules.additionalAggregationLabels }}{{ . }},{{ end }}cluster, namespace, pod, qos_class) (
+                      kube_pod_status_qos_class{job="kube-state-metrics"} > 0
+                    )
                 * on ({{ range $.Values.defaultRules.additionalAggregationLabels }}{{ . }},{{ end }}cluster, namespace, pod) group_left (ip)
                   max by ({{ range $.Values.defaultRules.additionalAggregationLabels }}{{ . }},{{ end }}cluster, namespace, pod, ip) (
-                    kube_pod_ips{job="kube-state-metrics"} > 0
+                      kube_pod_ips{job="kube-state-metrics"}
+                    or
+                        sum by ({{ range $.Values.defaultRules.additionalAggregationLabels }}{{ . }},{{ end }}pod, namespace, cluster) (kube_pod_info{job="kube-state-metrics"})
+                      unless
+                        sum by ({{ range $.Values.defaultRules.additionalAggregationLabels }}{{ . }},{{ end }}pod, namespace, cluster) (kube_pod_ips{job="kube-state-metrics"})
                   )
               * on ({{ range $.Values.defaultRules.additionalAggregationLabels }}{{ . }},{{ end }}cluster, namespace, pod) group_left (phase)
                 max by ({{ range $.Values.defaultRules.additionalAggregationLabels }}{{ . }},{{ end }}cluster, namespace, pod, phase) (kube_pod_status_phase{job="kube-state-metrics"} > 0)

ks-prometheus/components/wiztelemetry-mixin/rules/custom.libsonnet

@@ -26,14 +26,18 @@
             record: 'workspace_workload_node:kube_pod_info:',
             expr: |||
               max by (%(clusterLabel)s, node, ip, workspace, namespace, pod, qos_class, phase, workload, workload_type) (
-                        kube_pod_info{%(kubeStateMetricsSelector)s}
-                      * on (%(clusterLabel)s, namespace, pod) group_left (qos_class)
-                        max by (%(clusterLabel)s, namespace, pod, qos_class) (
-                          kube_pod_status_qos_class{%(kubeStateMetricsSelector)s} > 0
-                        )
+                          kube_pod_info{%(kubeStateMetricsSelector)s}
+                        * on (%(clusterLabel)s, namespace, pod) group_left (qos_class)
+                          max by (%(clusterLabel)s, namespace, pod, qos_class) (
+                            kube_pod_status_qos_class{%(kubeStateMetricsSelector)s} > 0
+                          )
                       * on (%(clusterLabel)s, namespace, pod) group_left (ip)
                         max by (%(clusterLabel)s, namespace, pod, ip) (
-                          kube_pod_ips{%(kubeStateMetricsSelector)s} > 0
+                            kube_pod_ips{%(kubeStateMetricsSelector)s}
+                          or
+                              sum by (pod, namespace, cluster) (kube_pod_info{%(kubeStateMetricsSelector)s})
+                            unless
+                              sum by (pod, namespace, cluster) (kube_pod_ips{%(kubeStateMetricsSelector)s})
                         )
                     * on (%(clusterLabel)s, namespace, pod) group_left (phase)
                       max by (%(clusterLabel)s, namespace, pod, phase) (kube_pod_status_phase{%(kubeStateMetricsSelector)s} > 0)

ks-prometheus/manifests/wiztelemetry-prometheusRule.yaml

@@ -14,14 +14,18 @@ spec:
     rules:
     - expr: |
         max by (cluster, node, ip, workspace, namespace, pod, qos_class, phase, workload, workload_type) (
-                  kube_pod_info{job="kube-state-metrics"}
-                * on (cluster, namespace, pod) group_left (qos_class)
-                  max by (cluster, namespace, pod, qos_class) (
-                    kube_pod_status_qos_class{job="kube-state-metrics"} > 0
-                  )
+                    kube_pod_info{job="kube-state-metrics"}
+                  * on (cluster, namespace, pod) group_left (qos_class)
+                    max by (cluster, namespace, pod, qos_class) (
+                      kube_pod_status_qos_class{job="kube-state-metrics"} > 0
+                    )
                 * on (cluster, namespace, pod) group_left (ip)
                   max by (cluster, namespace, pod, ip) (
-                    kube_pod_ips{job="kube-state-metrics"} > 0
+                      kube_pod_ips{job="kube-state-metrics"}
+                    or
+                        sum by (pod, namespace, cluster) (kube_pod_info{job="kube-state-metrics"})
+                      unless
+                        sum by (pod, namespace, cluster) (kube_pod_ips{job="kube-state-metrics"})
                   )
               * on (cluster, namespace, pod) group_left (phase)
                 max by (cluster, namespace, pod, phase) (kube_pod_status_phase{job="kube-state-metrics"} > 0)