Skip to content
Branch: master
Find file History
stefanprodan and jaypipes Fixes for App Mesh inject and Node termination charts (#44)
* Update App Mesh inject to 0.3.1
Fix for Jaeger tracing injection

Signed-off-by: stefanprodan <>

* Fix AWS Node termination chart
- use release namespace in cluster role binding
- set image tag to v1.0.0

Signed-off-by: stefanprodan <>

* Update Helm to v2.16.1 and v3.0.0

Signed-off-by: stefanprodan <>
Latest commit 57d641f Dec 2, 2019

AWS Node Termination Handler

AWS Node Termination Handler Helm chart for Kubernetes. For more information on this project see the project repo at


  • Kubernetes >= 1.11

Installing the Chart

Add the EKS repository to Helm:

helm repo add eks

Install AWS Node Termination Handler: To install the chart with the release name aws-node-termination-handler and default configuration:

helm install --name aws-node-termination-handler \
  --namespace kube-system eks/aws-node-termination-handler

To install into an EKS cluster where the Node Termination Handler is already installed, you can run:

helm upgrade --install --recreate-pods --force \
  aws-node-termination-handler --namespace kube-system eks/aws-node-termination-handler

If you receive an error similar to Error: release aws-node-termination-handler failed: <resource> "aws-node-termination-handler" already exists, simply rerun the above command.

The configuration section lists the parameters that can be configured during installation.

Uninstalling the Chart

To uninstall/delete the aws-node-termination-handler deployment:

helm delete --purge aws-node-termination-handler

The command removes all the Kubernetes components associated with the chart and deletes the release.


The following tables lists the configurable parameters of the chart and their default values.

Parameter Description Default
image.repository image repository amazon/aws-node-termination-handler
image.tag image tag <VERSION>
image.pullPolicy image pull policy IfNotPresent
deleteLocalData Tells kubectl to continue even if there are pods using emptyDir (local data that will be deleted when the node is drained). false
gracePeriod The time in seconds given to each pod to terminate gracefully. If negative, the default value specified in the pod will be used. 30
ignoreDaemonsSets Causes kubectl to skip daemon set managed pods true
affinity node/pod affinities None
podSecurityContext Pod Security Context {}
podAnnotations annotations to add to each pod {}
priorityClassName Name of the priorityClass system-node-critical
resources Resources for the pods requests.cpu: 50m, requests.memory: 64Mi, limits.cpu: 100m, limits.memory: 128Mi
securityContext Container Security context privileged: true
nodeSelector Tells the daemon set where to place the node-termination-handler pods. For example: lifecycle: "Ec2Spot", on-demand: "false", "spot", etc. Value must be a valid yaml expression. {}
tolerations list of node taints to tolerate []
rbac.create if true, create and use RBAC resources true
rbac.pspEnabled If true, create and use a restricted pod security policy false
serviceAccount.create If true, create a new service account true Service account to be used None
You can’t perform that action at this time.