As described in this this video, Sysdig Monitor now collects kube-state-metrics for monitoring and alerting. New dashboards provide out-of-the-box visibility of metrics for nodes, namespaces, services, daemonSets, jobs, replicaSets and pods.
To enable Kube State metrics, complete the following steps:
Step 1: Make sure you're running Agent version 0.77 or newer in a Daemonset-based Agent install.
NOTE: At this time, Kube State Metrics are only supported in a Daemonset-based install.
Step 2: Modify the Agent's YAML configuration to add the following top-level config flag:
In your Daemonset-based install, this can be done by specifying ADDITIONAL_CONF, such as in the bottom two lines of this example config snippet:
containers: - name: sysdig-agent image: sysdig/agent securityContext: privileged: true env: - name: ACCESS_KEY value: xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx - name: ADDITIONAL_CONF value: "dev_new_k8s: true\n"
Step 3: Activate the new configuration by restarting the Agent, such as by redeploying the Daemonset. For example:
kubectl delete daemonset sysdig-agent kubectl create -f sysdig-daemonset.yaml
Want to learn more about what you can do with this feature? Read our blog.