Wavefront Proxy Kustomization

This repository consolidates advanced configuration of Wavefront proxy through Kustomize. This allows easy reference of configuration parameters and deployment of a customised Wavefront proxy in Kubernetes. This is a good starting point to modify Wavefront proxy deployment to existing automation pipelines, especially if not using the Wavefront Helm chart.

Source: Wavefront Proxy Advanced Configuration Guide

Getting Started

This guide assumes that you are familiar with Kustomize and have a ready cluster (e.g. minikube) with kubectl to run/test the kustomizations.

• Install Kustomize on your machine
• Clone this repository and note the structure as follows:

base/ - adds wavefront.conf and log4j.xml advanced configuration to upstream K8s proxy manifest. Environment variables override configuration files.

overlays/memory-limit/ - overlays, or "inherits" base/ and limits memory use of Wavefront proxy by adjusting Java heap size and patching the K8s container resource limits.

overlays/tcpdump-sidecar - injects a tcpdump sidecar to the wavefront proxy pod for troubleshooting.

• Add your own WAVEFRONT_TOKEN and WAVEFRONT_URL credentials to base/kustomization.yaml

• To preview the generated manifest, run in base/ or memory-limit/ directory:

$ kustomize build

• To apply the generated manifest:

$ kustomize build | kubectl apply --namespace wavefront -f -

Alternatively, you may use built-in support for Kustomize: kubectl apply -k .

• To cleanup and remove the applied manifests, replace apply with delete

Note about secrets management

This kustomization does not integrate K8s secrets management. For non-trivial deployments, it is recommended to keep your Wavefront tokens safe. For more details, please visit Vault or SealedSecrets, depending on your environment and security needs.

Future Improvements

• Add native remote loading of upstream resources when this feature stabilises in Kustomize. 1, 2
• Add preprocessor rules configuration