Skip to content

setup-runtime-deploy.md

Latest commit

 

History

History
63 lines (43 loc) · 2.94 KB

setup-runtime-deploy.md

File metadata and controls

63 lines (43 loc) · 2.94 KB
Raw

Deploying the Mountain Passes Stack

Before we push the stack to our cluster with ArgoCD, we need to setup ArgoCD notifications to push deployment information back to Datadog.

The sample application includes notifications.yaml, which configures ArgoCD to notify Datadog of the status of deployments. As supplied, this applies to all applications deployed using ArgoCD in this cluster. For this to work you’ll need to add your Datadog API key to ArgoCD’s secret.

Trigger the deployment

To deploy the application, simply review and then apply the kustomize manifests from the argocd-manifests directory to your cluster. These use the ArgoCD CRDs to setup and deploy the images from your repository.

# Deploy  
kubectl apply -k argocd-manifests

# List ArgoCD applications and deployed pods   
kubectl get applications -A  
kubectl get pods -n sdlc-gitops-sample-stack

Important

if you’ve forked to a private repository, you’ll need to:

  • add docker secrets to your cluster to pull
  • add GitHub credentials to the ArgoCD UI to allow it to clone the repository

Wait for the stack to come up, and then forward a port to pass-summary-api:

 kubectl port-forward svc/pass-summary-api 8080:8080 \-n sdlc-gitops-sample-stack

Then, in another shell:

curl http://localhost:8080/pass-summary

Which should return something like {"total_ascent":6518,"pass_count":3}

Tip

k9s is a great terminal UI to work with Kubernetes clusters if you're looking for an alternative to kubectl.

The request here calls from the pass-summary-api, to the pass-api, which retrieves pass data from the database. This will result in log, trace and metric data being pushed to Datadog - let’s go have a look.

Find the deployment in Datadog

Tip

Datadog Continuous Delivery (CD) Visibility provides observability on your deployments.

Using the Go To dialog again, jump over to the Executions view from CD Visibility. Here you should immediately see the ArgoCD run that deployed the application into your cluster with each service broken out separately:

CD Visibility

Clicking onto the deployment for one of the services, we can see what triggered the deployment, who pushed the commit, and the associated commit hash:

CD Visibility - details

Clicking over to the Service tab shows us the metadata we provided earlier on for service ownership, and provides quick access to the logs/traces/metrics associated with the deployment in our running environment.

Last of all - let's go look at runtime visibility.