Supported OS
This check monitors Kube_metrics_server v0.3.0+, a component used by the Kubernetes control plane.
The Kube_metrics_server check is included in the Datadog Agent package. No additional installation is needed on your server.
To configure this check for an Agent running on a host:
Edit the kube_metrics_server.d/conf.yaml file, in the conf.d/ folder at the root of your Agent’s configuration directory to start collecting your kube_metrics_server performance data. See the sample kube_metrics_server.d/conf.yaml
for all available configuration options.
For containerized environments, see the Kubernetes Autodiscovery Integration Templates for guidance on applying the parameters below.
| Parameter | Value |
|---|---|
<INTEGRATION_NAME> | kube_metrics_server |
<INIT_CONFIG> | blank or {} |
<INSTANCE_CONFIG> | {"prometheus_url": "https://%%host%%:443/metrics"} |
If your endpoint is secured, additional configuration is required:
Identify the certificate used for securing the metric endpoint.
Mount the related certificate file in the Agent pod.
Apply your SSL configuration. See the default configuration file for more information.
Run the Agent’s status subcommand
and look for kube_metrics_server under the Checks section.
| kube_metrics_server.authenticated_user.requests (count) | Counter of authenticated requests broken out by username |
| kube_metrics_server.go.gc_duration_seconds.quantile (gauge) | GC invocation durations quantiles |
| kube_metrics_server.go.gc_duration_seconds.sum (gauge) | GC invocation durations sum |
| kube_metrics_server.go.gc_duration_seconds.count (gauge) | Number of the GC invocation |
| kube_metrics_server.go.goroutines (gauge) | Number of goroutines that currently exist |
| kube_metrics_server.kubelet_summary_request_duration.count (gauge) | Number of Kubelet summary request |
| kube_metrics_server.kubelet_summary_request_duration.sum (gauge) | The Kubelet summary request latencies sum |
| kube_metrics_server.kubelet_summary_scrapes_total (count) | Total number of attempted Summary API scrapes done by Metrics Server |
| kube_metrics_server.manager_tick_duration.count (gauge) | The total time spent collecting and storing metrics |
| kube_metrics_server.manager_tick_duration.sum (gauge) | The total time spent collecting and storing metrics |
| kube_metrics_server.scraper_duration.count (gauge) | Time spent scraping sources |
| kube_metrics_server.scraper_duration.sum (gauge) | Time spent scraping sources |
| kube_metrics_server.scraper_last_time (gauge) | Last time metrics-server performed a scrape since unix epoch |
| kube_metrics_server.process.max_fds (gauge) | Maximum number of open file descriptors |
| kube_metrics_server.process.open_fds (gauge) | Number of open file descriptors |
kube_metrics_server does not include any events.
kube_metrics_server.prometheus.health
Returns CRITICAL if the check cannot access the metrics endpoint.
Statuses: ok, critical
kube_metrics_server.up
Returns CRITICAL if Kubernetes Metrics Server is not healthy.
Statuses: ok, critical
Need help? Contact Datadog support .
