Install
helm install image-transformer charts/image-transformer -n image-transformer --create-namespace --set postgres.password=postgres
Verify
kubectl get pods,pvc,statefulsets -n image-transformer
kind doesn't see your local Docker daemon's images automatically — after any change to
backend/, rebuild the image and load it into the cluster before upgrading, otherwise the
running pod keeps the old code (builds are cache-fast, so just always rebuild):
docker build -t image-transformer-backend:local backend
kind load docker-image image-transformer-backend:local --name lab
Upgrading
helm upgrade image-transformer charts/image-transformer -n image-transformer
Verify s3 bucket + sqs
kubectl logs job/floci-bootstrap -n image-transformer
make_bucket: uploads
{
"QueueUrl": "http://localhost:4566/000000000000/image-processing"
}
Floci bootstrap complete: bucket=uploads queue=image-processing
for testing backend forward port to host
kubectl port-forward -n image-transformer svc/backend 8000:8000
processing image
kubectl port-forward -n image-transformer svc/backend 8000:8000
upload hero.png
curl -X POST http://localhost:8000/uploads -F "file=hero.png;type=image/png"
checking processor
kubectl logs -n image-transformer deploy/processor -f
download from bucket
kubectl port-forward -n image-transformer svc/floci 4566:4566
curl -o thumb.png "http://localhost:4566/uploads/thumbnails/<id>/hero.png"
docker build -t image-transformer-backend:local backend
docker build -t image-transformer-processor:local processor
docker build -t image-transformer-frontend:local frontend
kind load docker-image image-transformer-backend:local image-transformer-processor:local image-transformer-frontend:local --name lab
helm upgrade image-transformer charts/image-transformer -n image-transformer
kubectl port-forward -n image-transformer svc/frontend 8080:80
The kind cluster needs to be created with extraPortMappings for 80/443 and the
ingress-ready=true node label before ingress-nginx will work — recreate it with the
committed config (this deletes and rebuilds the cluster, so any images loaded into it
will need kind load docker-image again):
kind delete cluster --name lab
kind create cluster --name lab --config kind-config.yaml
Install ingress-nginx (kind-specific manifest — deploys the controller with hostPort
80/443 on the node labeled ingress-ready=true):
kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/main/deploy/static/provider/kind/deploy.yaml
It is required to set nodeSelector: {ingress-ready: "true"} on the
controller Deployment, so it can land on a worker node with 80/443
extraPortMappings from kind-config.yaml. Force it onto
the control-plane node:
kubectl patch deployment ingress-nginx-controller -n ingress-nginx --type=json \
-p='[{"op":"add","path":"/spec/template/spec/nodeSelector/ingress-ready","value":"true"}]'
kubectl wait --namespace ingress-nginx --for=condition=ready pod --selector=app.kubernetes.io/component=controller --timeout=180s
Rebuild the app images, load them into the fresh cluster, and install the release (it's a new
cluster so helm install,):
docker build -t image-transformer-backend:local backend
docker build -t image-transformer-processor:local processor
docker build -t image-transformer-frontend:local frontend
kind load docker-image image-transformer-backend:local image-transformer-processor:local image-transformer-frontend:local --name lab
helm install image-transformer charts/image-transformer -n image-transformer --create-namespace --set postgres.password=postgres
kubectl get pods,ingress -n image-transformer
Test in the browser at http://localhost/ — the frontend is served at /, and its
uploads go through http://localhost/api/uploads (routed straight to the backend
Service by the ingress, bypassing the frontend's own nginx /api proxy).