newrelic-k8s-metrics-adapter

A Helm chart to deploy the New Relic Kubernetes Metrics Adapter.

Homepage: https://hub.docker.com/r/newrelic/newrelic-k8s-metrics-adapter

Source Code

• https://github.com/newrelic/newrelic-k8s-metrics-adapter
• https://github.com/newrelic/newrelic-k8s-metrics-adapter/tree/main/charts/newrelic-k8s-metrics-adapter

Requirements

Repository	Name	Version
https://helm-charts.newrelic.com	common-library	1.2.0

Values

Key	Type	Default	Description
affinity	object	{}	Node affinity to use for scheduling.
apiServicePatchJob.image	object	See values.yaml.	Registry, repository, tag, and pull policy for the job container image.
apiServicePatchJob.volumeMounts	list	[]	Additional Volume mounts for Cert Job, you might want to mount tmp if Pod Security Policies.
apiServicePatchJob.volumes	list	[]	Additional Volumes for Cert Job.
certManager.enabled	bool	false	Use cert manager for APIService certs, rather than the built-in patch job.
config.accountID	string	nil	New Relic Account ID where the configured metrics are sourced from. (Required)
config.cacheTTLSeconds	int	30	Period of time in seconds in which a cached value of a metric is consider valid.
config.externalMetrics	string	See values.yaml	Contains all the external metrics definition of the adapter. Each key of the externalMetric entry represents the metric name and contains the parameters that defines it.
config.nrdbClientTimeoutSeconds	int	30	Defines the NRDB client timeout. The maximum allowed value is 120.
config.region	string	Automatically detected from licenseKey.	New Relic account region. If not set, it will be automatically derived from the License Key.
containerSecurityContext	string	nil	Configure containerSecurityContext
extraEnv	list	[]	Array to add extra environment variables
extraEnvFrom	list	[]	Array to add extra envFrom
extraVolumeMounts	list	[]	Add extra volume mounts
extraVolumes	list	[]	Array to add extra volumes
fullnameOverride	string	""	To fully override common.naming.fullname
image	object	See values.yaml.	Registry, repository, tag, and pull policy for the container image.
image.pullSecrets	list	[]	The image pull secrets.
nodeSelector	object	{}	Node label to use for scheduling.
personalAPIKey	string	nil	New Relic Personal API Key (stored in a secret). Used to connect to NerdGraph in order to fetch the configured metrics. (Required)
podAnnotations	string	nil	Additional annotations to apply to the pod(s).
podSecurityContext	string	nil	Configure podSecurityContext
proxy	string	nil	Configure proxy for the metrics-adapter.
rbac.pspEnabled	bool	false	Whether the chart should create Pod Security Policy objects.
replicas	int	1	Number of replicas in the deployment.
resources	object	See values.yaml	Resources you wish to assign to the pod.
serviceAccount.create	string	true	Specifies whether a ServiceAccount should be created for the job and the deployment. false avoids creation, true or empty will create the ServiceAccount
serviceAccount.name	string	Automatically generated.	If serviceAccount.create this will be the name of the ServiceAccount to use. If not set and create is true, a name is generated using the fullname template. If create is false, a serviceAccount with the given name must exist
tolerations	list	[]	List of node taints to tolerate (requires Kubernetes >= 1.6)
verboseLog	bool	false	Enable metrics adapter verbose logs.

Example

Make sure you have added the New Relic chart repository.

Because of metrics configuration, we recommend to use an external values file to deploy the chart. An example with the required parameters looks like:

cluster: ClusterName
personalAPIKey: <Personal API Key>
config:
  accountID: <Account ID>
  externalMetrics:
    nginx_average_requests:
      query: "FROM Metric SELECT average(nginx.server.net.requestsPerSecond) SINCE 2 MINUTES AGO"

Then, to install this chart, run the following command:

helm upgrade --install [release-name] newrelic-k8s-metrics-adapter/newrelic-k8s-metrics-adapter --values [values file path]

Once deployed the metric nginx_average_requests will be available to use by any HPA. This is and example of an HPA yaml using this metric:

kind: HorizontalPodAutoscaler
apiVersion: autoscaling/v2beta2
metadata:
  name: nginx-scaler
spec:
  scaleTargetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: nginx
  minReplicas: 1
  maxReplicas: 10
  metrics:
    - type: External
      external:
        metric:
          name: nginx_average_requests
          selector:
            matchLabels:
              k8s.namespaceName: nginx
        target:
          type: Value
          value: 10000

The NRQL query that will be run to get the nginx_average_requests value will be:

FROM Metric SELECT average(nginx.server.net.requestsPerSecond) WHERE clusterName='ClusterName' AND `k8s.namespaceName`='nginx' SINCE 2 MINUTES AGO

External Metrics

An example of multiple external metrics defined:

externalMetrics:
    nginx_average_requests:
      query: "FROM Metric SELECT average(nginx.server.net.requestsPerSecond) SINCE 2 MINUTES AGO"
    container_average_cores_utilization:
      query: "FROM Metric SELECT average(`k8s.container.cpuCoresUtilization`) SINCE 2 MINUTES AGO"

Resources

The default set of resources assigned to the newrelic-k8s-metrics-adapter pods is shown below:

resources:
  limits:
    memory: 80M
  requests:
    cpu: 100m
    memory: 30M

Maintainers

• juanjjaramillo
• csongnr
• dbudziwojskiNR