Workspace Cluster

User workspaces are deploying with an ephemeral, namespaced Kubernetes cluster. If you need to interact with this cluster...

If you prefer video form, click here for a 5min video!

For more information about DevZero Kubernetes Clusters, visit Kubernetes.

From Anywhere

To view the Kubernetes configuration for a workspace, run dz workspace kubeconfig $workspace_name

For example, if you have a workspace named "big-skylark-sezs"

dz workspace kubeconfig big-skylark-sezs

To write the config to the default Kubernetes configuration location, run

For example, if you have a workspace named "big-skylark-sezs"

dz workspace kubeconfig big-skylark-sezs --update-kubeconfig

Then run commands like

kubectl get pods

From Inside a DevBox

When inside a DevBox context (i.e. when connected to a workspace), the CLI is able to retrieve environmental information from /etc/devzero.

In a fresh new workspace, you can immediately run:

kubectl get pods

DevZero reserves the default namespace for it's managed deployments. Do not operate on this namespace. Adding or removing resources in this namespace will lead to undocumented behaviors and cause potential data loss.

dz workspace kubeconfig -h
This kubeconfig can be used by any kubectl to interact with the virtual cluster backing a workspace.
Usage: kubectl --kubeconfig <(dz ws kubeconfig <workspace_id | workspace_name>) ...

Usage:
  dz workspace kubeconfig <workspace_id | workspace_name> [flags]

Aliases:
  kubeconfig, kc

Flags:
  -h, --help                help for kubeconfig
  -u, --update-kubeconfig   update local kubeconfig (default: $HOME/.kube/config)

Global Flags:
      --verbose   Get detailed output

Tutorial Video

Here's a video covering how you can access and deploy apps to your workspace's Kubernetes cluster...

Tutorial Steps

Create a recipe that you can use at devzero.io/dashboard/recipes/new (give it any name and leave everything else blank and click Create a recipe).

Use the following recipe, then Save and Build and then Launch once the build completes successfully (it uses Google Cloud Platform's microservices-demo repo).

version: "3"
build:
  steps:
    - type: apt-get
      packages:
        [
          "apt-transport-https",
          "build-essential",
          "ca-certificates",
          "curl",
          "git",
          "nano",
          "software-properties-common",
          "ssh",
          "sudo",
          "tar",
          "unzip",
          "vim",
          "wget",
          "zip",
        ]
    - type: git-clone
      url: https://github.com/GoogleCloudPlatform/microservices-demo
    - type: command
      command: |
        curl -LO "https://dl.k8s.io/release/$(curl -L -s https://dl.k8s.io/release/stable.txt)/bin/linux/amd64/kubectl"
        sudo install -o root -g root -m 0755 kubectl /usr/local/bin/kubectl && rm kubectl
    - type: command
      command: |
        curl -Lo skaffold https://storage.googleapis.com/skaffold/releases/latest/skaffold-linux-amd64 && \
        sudo install skaffold /usr/local/bin/
    - type: apt-get
      packages: ["docker-ce", "docker-ce-cli", "containerd.io"]
      extra_repositories:
        - key_url: https://download.docker.com/linux/ubuntu/gpg
          repository: https://download.docker.com/linux/ubuntu
          components: []
          distribution: ""
    - type: command
      command: |
        usermod -aG docker devzero
        systemctl enable docker.service
        systemctl enable containerd.service
      user: root

Build a workspace from the recipe, and run the following in your terminal:

dz workspace connect $workspace_name

Run the following steps inside the SSH session that's connected to your workspace:

kubectl get pods  # verification
cd /home/devzero/microservices-demo
skaffold run --default-repo ttl.sh  # this will take a bit of time since its building multiple docker images (~5mins)

Verify that all the pods are running:

kubectl get pods

Port forward the frontend to your terminal session

kubectl port-forward --address 0.0.0.0 deployment/frontend 8088:8080

Visit, $workspace_name:8088 in your browser to see the app!

Workspace Cluster

From Anywhere

From Inside a DevBox

Tutorial Video

On this page