aws-vpc-cni

AWS VPC CNI

This chart installs the AWS CNI Daemonset: https://github.com/aws/amazon-vpc-cni-k8s

Prerequisites

• Kubernetes 1.11+ running on AWS

Installing the Chart

First add the EKS repository to Helm:

helm repo add eks https://aws.github.io/eks-charts

To install the chart with the release name aws-vpc-cni and default configuration:

$ helm install --name aws-vpc-cni --namespace kube-system eks/aws-vpc-cni

To install into an EKS cluster where the CNI is already installed, see this section below

Configuration

The following table lists the configurable parameters for this chart and their default values.

Parameter	Description	Default
affinity	Map of node/pod affinities	{}
cniConfig.enabled	Enable overriding the default 10-aws.conflist file	false
cniConfig.fileContents	The contents of the custom cni config file	nil
eniConfig.create	Specifies whether to create ENIConfig resource(s)	false
eniConfig.region	Region to use when generating ENIConfig resource names	us-west-2
eniConfig.subnets	A map of AZ identifiers to config per AZ	nil
eniConfig.subnets.id	The ID of the subnet within the AZ which will be used in the ENIConfig	nil
eniConfig.subnets.securityGroups	The IDs of the security groups which will be used in the ENIConfig	nil
env	List of environment variables. See here for options	(see values.yaml)
fullnameOverride	Override the fullname of the chart	aws-node
image.region	ECR repository region to use. Should match your cluster	us-west-2
image.tag	Image tag	v1.7.5
image.pullPolicy	Container pull policy	IfNotPresent
image.override	A custom docker image to use	nil
imagePullSecrets	Docker registry pull secret	[]
init.image.region	ECR repository region to use. Should match your cluster	us-west-2
init.image.tag	Image tag	v1.7.5
init.image.pullPolicy	Container pull policy	IfNotPresent
init.image.override	A custom docker image to use	nil
init.env	List of init container environment variables. See here for options	(see values.yaml)
init.securityContext	Init container Security context	privileged: true
originalMatchLabels	Use the original daemonset matchLabels	false
nameOverride	Override the name of the chart	aws-node
nodeSelector	Node labels for pod assignment	{}
podSecurityContext	Pod Security Context	{}
podAnnotations	annotations to add to each pod	{}
podLabels	Labels to add to each pod	{}
priorityClassName	Name of the priorityClass	system-node-critical
resources	Resources for the pods	requests.cpu: 10m
securityContext	Container Security context	capabilities: add: - "NET_ADMIN"
serviceAccount.name	The name of the ServiceAccount to use	nil
serviceAccount.create	Specifies whether a ServiceAccount should be created	true
serviceAccount.annotations	Specifies the annotations for ServiceAccount	{}
livenessProbe	Livenness probe settings for daemonset	(see values.yaml)
readinessProbe	Readiness probe settings for daemonset	(see values.yaml)
crd.create	Specifies whether to create the VPC-CNI CRD	true
tolerations	Optional deployment tolerations	[]
updateStrategy	Optional update strategy	type: RollingUpdate

Specify each parameter using the --set key=value[,key=value] argument to helm install or provide a YAML file containing the values for the above parameters:

$ helm install --name aws-vpc-cni --namespace kube-system eks/aws-vpc-cni --values values.yaml

Adopting the existing aws-node resources in an EKS cluster

If you do not want to delete the existing aws-node resources in your cluster that run the aws-vpc-cni and then install this helm chart, you can adopt the resources into a release instead. This process is highlighted in this PR comment. Once you have annotated and labeled all the resources this chart specifies, enable the originalMatchLabels flag, and also set crd.create to false on the helm release and run an update. If you have been careful this should not diff and leave all the resources unmodified and now under management of helm.

Here is an example script to modify the existing resources:

WARNING: Substitute YOUR_HELM_RELEASE_NAME_HERE with the name of your helm release.

#!/usr/bin/env bash

set -euo pipefail

# don't import the crd. Helm cant manage the lifecycle of it anyway.
for kind in daemonSet clusterRole clusterRoleBinding serviceAccount; do
  echo "setting annotations and labels on $kind/aws-node"
  kubectl -n kube-system annotate --overwrite $kind aws-node meta.helm.sh/release-name=YOUR_HELM_RELEASE_NAME_HERE
  kubectl -n kube-system annotate --overwrite $kind aws-node meta.helm.sh/release-namespace=kube-system
  kubectl -n kube-system label --overwrite $kind aws-node app.kubernetes.io/managed-by=Helm
done