Skip to content

Latest commit

 

History

History
109 lines (75 loc) · 4.81 KB

readme.md

File metadata and controls

109 lines (75 loc) · 4.81 KB
Raw

lab-01 - provision workshop resources

Estimated completion time - 20 min

As always, to start working with labs, we need to provision Azure Kubernetes Service (AKS) instance. For this workshop we will use Bicep. AKS and supporting resources are not free and the compute power will come with some costs, but we will use the smallest Virtual Machine size for our nodes and we will use just a few nodes when we will work with cluster autoscaling. Also, remember to delete all resources when you are finished working with the labs. Here is the list of resources we need to provision:

  • Resource Group
  • Azure Container Registry (ACR)
  • Azure Log Analytics
  • Azure Kubernetes Service (AKS)

Our AKS cluster will fullfil the following requirements:

  • Use managed identities in AKS to create additional resources like load balancers and managed disks in Azure
  • Integrate AKS with Azure Log Analytics for monitoring
  • Integrate AKS with Azure Container Registry

Here is the complete visualization of resources we will provision.

model

Goals

  • Provision workshop resources

Task #1 - deploy workshop resources

All Bicep templates are located under infra folder and are split into the following modules:

Module name Description
acr.bicep Azure Container Registry resource implementation
vnet.bicep Private Virtual Network implementation
logAnalytics.bicep Log Analytics implementation
aks.bicep AKS implementation
attachACRToAKS.bicep Contains code that attaches ACR to AKS instance
grantAksPermissions.bicep Contains code that assign Network Contributor Role to AKS principal id

Deployment is orchestrated by the deployment.bicep template. All parameters are extracted into parameters-blue.json file.

Let's provision blue cluster first.

# Select your subscription
az account set --subscription <YOUR-SUBSCRIPTION-ID>

# Deploy workshop resources
az deployment sub create --location westeurope --template-file ./deployment.bicep  --parameters './parameters.json'
 \ Running ..

# When provisioned (it takes approx. 5 min), connect to your cluster
az aks get-credentials --resource-group iac-ws5-rg --name iac-ws5-aks --overwrite-existing

# Install kubectl
az aks install-cli

If you use PowerShell, you need to update system PATH environment variable and add new item for %userprofile%\.azure-kubectl.

env

  • Open the Start Search, type in env, and choose Edit the system environment variables
  • Click the Environment Variables… button.
  • Select Path variable under System variables section
  • CLick Edit...
  • Click New and set variable to %userprofile%\.azure-kubectl

You need to reset your PowerShell (and cmd) session(s) for change to take effect.

If you are running on WSL, you may need to use sudo command, in this case run

# Install kubectl using sudo
sudo az aks install-cli

When kubectl is installed, test it by getting list of namespaces...

# Get list of namespaces
kubectl get ns
NAME              STATUS   AGE
calico-system     Active   3m51s
default           Active   6m21s
kube-node-lease   Active   6m24s
kube-public       Active   6m24s
kube-system       Active   6m24s
tigera-operator   Active   6m

If you use Oh My Posh, you can configure that the name of active cluster is shown at the command line prompt, as it's shown below:

k8s-at-the-command-line

Learn how to setup your shell (bash or PowerShell) for better AKS/kubectl experience

Useful links

Next: deploy test application

Go to lab-02