Deploying Our Legacy Applications to Kubernetes
The workshop already has an application ready to be deployed to kubernetes included.
It's a Python and Node microservices application. Today, we'll be focusing on replacing one of the pieces with a Golang service.
So let's get the Kubernetes cluster up and running, along with Datadog as a DaemonSet.
First, ensure your kubernetes cluster has been initialized, and both nodes have been added. You can do this by running
kubectl get nodes in host1.
On host1, change into the YAML file directory with a
cd k8s-yaml-files. You should be able to
ls and see the YAML files for every service we plan on running in our cluster.
First, add your Datadog API key to the secrets. You can do this with a:
$ kubectl create secret generic datadog-api --from-literal=token=<YOUR_DATADOG_API_KEY>
Once that's done, we'll next need to create a secret username and password for our PostgreSQL database.
$ kubectl create secret generic postgres-user --from-literal=token=postgres $ kubectl create secret generic postgres-password --from-literal=token=<YOUR_PASSWORD>
Finally, spin up the Datadog Agent container, so we can see our Nodes in the Datadog app:
$ cd k8s-yaml-files/ $ kubectl apply -f datadog-agent.yaml
You should now be able to see your cluster inside of Datadog. Verify this before continuing on to the next step.