Skip to content

Debugging a Remote ArgoCD Environment

In this guide, we will describe how to debug a remote ArgoCD environment with Telepresence.

Telepresence allows you to connect & debug a service deployed in a remote environment and to "cherry-pick" one service to run locally, staying connected to the remote cluster. This will:

  • Reduce resource footprint on the local machine
  • Decrease the feedback loop time
  • Result in more confidence about the delivered code.

To read more about it, refer to the official documentation at or Medium.

Install ArgoCD

First of all, install ArgoCD on your cluster

kubectl create ns argocd
curl -sSfL | kubectl apply -n argocd -f -


Connect to one of the services, for example, to debug the main ArgoCD server run:

telepresence --swap-deployment argocd-server --namespace argocd --env-file .envrc.remote --expose 8080:8080 --expose 8083:8083 --run bash
  • --swap-deployment changes the argocd-server deployment
  • --expose forwards traffic of remote ports 8080 and 8083 to the same ports locally
  • --env-file writes all the environment variables of the remote pod into a local file, the variables are also set on the subprocess of the --run command
  • --run defines which command to run once a connection is established, use bash, zsh or others


Once a connection is established, use your favorite tools to start the server locally.


  • Compile make server
  • Run ./dist/argocd-server


In VSCode use the integrated terminal to run the Telepresence command to connect. Then, to run argocd-server service use the following configuration. Update the configuration file to point to kubeconfig file: KUBECONFIG= (required)

            "name": "Launch",
            "type": "go",
            "request": "launch",
            "mode": "auto",
            "program": "${workspaceFolder}/cmd/argocd-server",
            "envFile": [
            "env": {
                "CGO_ENABLED": "0",
                "KUBECONFIG": "/path/to/kube/config"