Duplo-JIT + Lens Setup and Usage Guide

For Windows and macOS

---

🧩 What is Duplo-JIT?

duplo-jit (Just-In-Time access) is a CLI tool that:

• Authenticates you temporarily with Duplo.

• Generates a short-lived kubeconfig so you can securely connect to your Duplo-managed Kubernetes cluster.

---

⚙️ Installation Guide

🪟 For Windows

1. Download binaries from:
   🔗 https://github.com/duplocloud/duplo-jit/releases

2. Create a folder for Duplo-JIT:

    mkdir "C:\Program Files\duplo-jit"
   

3. Move binaries to the folder:

    move C:\Users\<you>\Downloads\duplo-jit.exe "C:\Program Files\duplo-jit\"
    move C:\Users\<you>\Downloads\duplo-aws-credential-process.exe "C:\Program Files\duplo-jit\"
   

4. Add the folder to your PATH:

    setx PATH "$($env:PATH);C:\Program Files\duplo-jit"
   

5. Or add it manually via GUI:

   • Search “Environment Variables” in Windows.

   • Click Edit the system environment variables → Environment Variables…

   • Under System variables, select Path → click Edit → New → add:

       C:\Program Files\duplo-jit
     

   • Click OK → OK → OK.

   • Restart PowerShell.

6. Verify:

    duplo-jit --version
   

✅ Example Output:

duplo-jit.exe version 0.5.11 (git commit 0818367...)

---

🍎 For macOS

1. Download the macOS binary (duplo-jit_0.x.x_darwin_amd64.tar.gz or arm64 for M1/M2 Macs) from:
   🔗 https://github.com/duplocloud/duplo-jit/releases

2. Extract it:

    tar -xvzf duplo-jit_0.5.11_darwin_amd64.tar.gz
   

3. Move it to a system path:

    sudo mv duplo-jit /usr/local/bin/
    sudo mv duplo-aws-credential-process /usr/local/bin/
   

4. Make sure both are executable:

    chmod +x /usr/local/bin/duplo-jit
    chmod +x /usr/local/bin/duplo-aws-credential-process
   

5. Verify installation:

    duplo-jit --version
   

✅ Example Output:

duplo-jit version 0.5.11 (git commit 0818367...)

---

🔐 Authenticate and Generate Kubeconfig

1. Login to Duplo:

    duplo-jit login
   

   This will open a browser window for authentication.

2. Fetch kubeconfig for your tenant (replace <tenant-name>):

    duplo-jit kubeconfig get <tenant-name>
   

3. This generates a file like:

    ~/Downloads/duploinfra-ense-use2-kubeconfig.yaml
   

---

💻 Connect to Your Cluster

Temporary connection:

kubectl get nodes --kubeconfig ~/Downloads/duploinfra-ense-use2-kubeconfig.yaml

Set as your default kubeconfig:

Windows:

setx KUBECONFIG "C:\Users\<you>\Downloads\duploinfra-ense-use2-kubeconfig.yaml"

macOS:

export KUBECONFIG=~/Downloads/duploinfra-ense-use2-kubeconfig.yaml

Then test:

kubectl get nodes

---

🧰 Using Lens (Kubernetes IDE)

1. Open Lens.

2. Go to Catalog → Clusters → Add Cluster.

3. Choose Add from kubeconfig.

4. Browse to and select your Duplo kubeconfig file.

5. Once added, click the cluster to connect.

⚠️ When access expires (usually after a few hours):

duplo-jit kubeconfig get <tenant-name>

Then reload the cluster in Lens.

---

🧭 Working with Multiple Clusters (e.g., Duplo + Minikube)

If you want to use Duplo and Minikube together:

1. Merge your kubeconfigs:

    export KUBECONFIG=~/.kube/config:~/Downloads/duploinfra-ense-use2-kubeconfig.yaml
    kubectl config view --merge --flatten > ~/.kube/config-merged
    mv ~/.kube/config-merged ~/.kube/config
   

2. Verify contexts:

    kubectl config get-contexts
   

3. Switch between clusters:

    kubectl config use-context <context-name>
   

4. Lens will automatically detect both clusters.

---

⏳ When Duplo-JIT Access Expires

If you see an authentication error like:

Error while proxying request: exec: executable duplo-jit not found

Run:

duplo-jit kubectl

This re-authenticates and refreshes your kubeconfig.

---

⚠️ Common Errors and Fixes

Error:

exec: executable duplo-jit not found

Cause: Duplo-JIT not in PATH.

Fix:

• Verify installation directory is in PATH.

• Restart terminal or Lens.

---

✅ You’re all set!
You can now:

• Use kubectl normally with your Duplo-managed cluster.

• Connect and manage clusters via Lens.

• Re-authenticate easily when access expires.

• Merge clusters like Minikube + Duplo for multi-cluster management.