Chapter 3. Enabling observability in Connectivity Link
Enabling observability for Connectivity Link components configures integration between Connectivity Link components, including any Gateways, and the Prometheus Operator installed in your OpenShift cluster when user monitoring is enabled.
This feature works by creating a set of ServiceMonitors and PodMonitors, which instruct Prometheus to scrape metrics from Connectivity Link and Gateway components. These scraped metrics are then used in the Connectivity Link example dashboards and alerts.
You must perform these steps on each OpenShift cluster that you want to use Connectivity Link on.
Prerequisites
- You have configured metrics as described in Chapter 2, Configuring observability metrics.
Procedure
To enable observability for Kuadrant and any Gateways, set
enable: truein theobservabilitysection in yourKuadrantcustom resource as follows:Copy to Clipboard Copied! Toggle word wrap Toggle overflow When enabled, Connectivity Link creates ServiceMonitors and PodMonitors for its components in the Connectivity Link Operator namespace. A single set of monitors is also created in each Gateway namespace to scrape metrics from any Gateways, and in the corresponding Gateway system namespace, in this case, the
istio-systemnamespace.Check the created monitors as follows:
kubectl get servicemonitor,podmonitor -A -l kuadrant.io/observability=true
kubectl get servicemonitor,podmonitor -A -l kuadrant.io/observability=trueCopy to Clipboard Copied! Toggle word wrap Toggle overflow NoteYou can make changes to monitors after they are created if needed. Monitors are only ever created or deleted, and not updated or reverted. If you decide the default monitors are not suitable, you can set
enable: falseand create your ownServiceMonitororPodMonitordefinitions, or configure Prometheus directly.