Wordsmith is the demo project originally shown at DockerCon EU 2017 and 2018.
The demo app runs across three containers:
- api - a Java REST API which serves words read from the database
- web - a Go web application that calls the API and builds words into sentences
- db - a Postgres database that stores words
The only requirement to build and run the app from source is Docker. Clone this repo and use Docker Compose to build all the images. You can use the new V2 Compose with docker compose or the classic docker-compose CLI:
docker compose up --buildOr you can pull pre-built images from Docker Hub using docker compose pull.
You can deploy the same app to Kubernetes using the Kustomize configuration. It will define all of the necessary Deployment and Service objects and a ConfigMap to provide the database schema.
Apply the manifest using kubectl while at the root of the project:
kubectl apply -k .Once the pods are running, browse to http://localhost:8080 and you will see the site.
Docker Desktop includes Kubernetes and the kubectl command line, so you can work directly with the cluster. Check the services are up, and you should see output like this:
kubectl get svc
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
db ClusterIP None <none> 55555/TCP 2m
kubernetes ClusterIP 10.96.0.1 <none> 443/TCP 38d
web LoadBalancer 10.107.215.211 <pending> 8080:30220/TCP 2m
words ClusterIP None <none> 55555/TCP 2m
Check the pods are running and you should see one pod each for the database and web components and five pods for the words API:
kubectl get pods
NAME READY STATUS RESTARTS AGE
db-8678676c79-h2d99 1/1 Running 0 1m
web-5d6bfbbd8b-6zbl8 1/1 Running 0 1m
api-858f6678-6c8kk 1/1 Running 0 1m
api-858f6678-7bqbv 1/1 Running 0 1m
api-858f6678-fjdws 1/1 Running 0 1m
api-858f6678-rrr8c 1/1 Running 0 1m
api-858f6678-x9zqh 1/1 Running 0 1m
We can make it even better by implementing ArgoCD, it's inspired by the best practices mentioned in this blog post
To install ArgoCD, follow the official installation guide. ArgoCD is a declarative, GitOps continuous delivery tool for Kubernetes.
To utilize ArgoCD in all it's capabilities let's use the 3-level structure as shown in the image below
At the lowest level, keep all the k8s yaml/helm. One level above we have Application Set which wraps all the k8s manifest into ArgoCd apps. This should be sufficient to go for now. The next level is optional, but it allows you to bootstrap all the apps in the cluster.
To deploy all the apps to the cluster, run the below,
kubectl apply -f root-argocd-app.yaml
You can also deploy apps based on env, for example, to deploy apps to dev
kubectl apply -f argo-appsets/dev-appset.yaml
With Argo we deploy the same set of apps to 3 environments dev, stage and prod. Base/common settings of the apps go to base and env specific configs go under envs.
git clone https://github.com/govindkailas/wordsmith.git
cd wordsmith
# kustomize
kustomize edit set image wordsmith-api:v2.0
# plain yaml
kubectl patch --local -f apps/api/deploy-svc.yaml -p '{"spec":{"template":{"spec":{"containers":[{"name":"api","image":"dockersamples/wordsmith-api:v2.0"}]}}}}' -o yaml
git add . -m "Update api to v2.0"
git push
After this ArgoCD will detect the change and auto sync the env based on your configuration.
You can also control the app you want to deploy based on envs under the apps. For example, api is deployed on all envs and web is deployed only in stage and prod. All 3 apps are deployed on prod env. Prod has its own configs that are differtent from other envs, like replicas of api are 5, resoruces allocated to each containers are more and web service is exposed on port#80. All of these are picked up based on the env file of prod from here
