Merge pull request #1442 from hsolberg/feature/make-honorLabels-configurable

Helm: Add option to configure honorLabels for serviceMonitor

Author: stefanprodan

charts/flagger/README.md

@@ -141,6 +141,7 @@ The following tables lists the configurable parameters of the Flagger chart and
 | `prometheus.retention`               | Prometheus data retention                                                                                                                          | `2h`                                  |
 | `selectorLabels`                     | List of labels that Flagger uses to create pod selectors                                                                                           | `app,name,app.kubernetes.io/name`     |
 | `serviceMonitor.enabled`             | If `true`, creates service and serviceMonitor for monitoring Flagger metrics                                                                                           | `false`     |
+| `serviceMonitor.honorLabels`         | If `true`, label conflicts are resolved by keeping label values from the scraped data and ignoring the conflicting server-side labels              | `false`                               |
 | `configTracking.enabled`             | If `true`, flagger will track changes in Secrets and ConfigMaps referenced in the target deployment                                                | `true`                                |
 | `eventWebhook`                       | If set, Flagger will publish events to the given webhook                                                                                           | None                                  |
 | `slack.url`                          | Slack incoming webhook                                                                                                                             | None                                  |

charts/flagger/templates/servicemonitor.yaml

@@ -13,7 +13,7 @@ spec:
       port: http
       interval: 30s
       scrapeTimeout: 30s
-      honorLabels: false
+      honorLabels: {{ .Values.serviceMonitor.honorLabels }}
   namespaceSelector:
     matchNames:
       - {{ .Release.Namespace }}

charts/flagger/values.yaml

@@ -26,6 +26,7 @@ metricsServer: "http://prometheus:9090"
 # creates serviceMonitor for monitoring Flagger metrics
 serviceMonitor:
   enabled: false
+  honorLabels: false
 
 # accepted values are kubernetes, istio, linkerd, appmesh, contour, nginx, gloo, skipper, traefik, apisix, osm
 meshProvider: ""