Chapter 2. Configuring observability metrics

Procedure

1. When monitoring for user projects is enabled in OpenShift, this allows the scraping of metrics from the Gateway and Connectivity Link components. To scrape additional metrics in OpenShift, you can install a kube-state-metrics instance, with a custom resource configuration as follows:

   kubectl apply -f https://raw.githubusercontent.com/Kuadrant/kuadrant-operator/v1.0.1/config/observability/openshift/kube-state-metrics.yaml
   
   kubectl apply -k https://github.com/Kuadrant/gateway-api-state-metrics/config/kuadrant?ref=0.6.0

   Copy to Clipboard

2. Some example dashboards aggregate data by request path. By default, OpenShift Service Mesh metrics do not include labels for request paths, however, you can enable these labels as follows:

   kubectl apply -f https://raw.githubusercontent.com/Kuadrant/kuadrant-operator/v1.0.1/config/observability/openshift/telemetry.yaml

   Copy to Clipboard
   Note

   This might lead to a high cardinality label where multiple time series are generated, which can have an impact on performance and resource usage. For more details, see the Reliable Insights blog post on cardinality.

3. You can configure scraping of metrics from the various Connectivity Link Operators as follows:

   kubectl apply -f https://raw.githubusercontent.com/Kuadrant/kuadrant-operator/v1.0.1/config/observability/prometheus/monitors/operators.yaml

   Copy to Clipboard
   Note

   You must apply additional configuration so that all relevant metrics are scraped. However, this configuration depends on how you deploy your Gateway later. For details on the required steps, see Configuring and deploying Gateway policies with Connectivity Link.