C4W2-Lab 2 Issue - Unable to access the deployment

Course Q&A Machine Learning Engineering for Production(MLOps) Deploying Machine Learning Models in Production
1

Hello, I followed the Lab 2 instructions below and try to access the deployment with docker as VM runtime

image
image1096×318 16 KB

But I cannot access and received the following message:
“! Because you are using a Docker driver on windows, the terminal needs to be open to run it.”

image
image1277×407 21.8 KB

image
image1261×674 88.2 KB

1 Like
2

I ran into the same issue with using minikube service tf-serving-service. I ended up getting it to work by switching to using LoadBalancer access. See Create an External Load Balancer | Kubernetes and Accessing apps | minikube. You can connect via an external IP after running minikube tunnel

3

Hello can you elaborate a little more on how you use the loadbalancer instead? I’ve changed the yaml in order to use the LoadBalancer type, but accessing the external IP now yields a ‘404 not found’.

4

This is what my service looks like:

$ kubectl get svc tf-serving-service
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
tf-serving-service LoadBalancer 10.100.92.73 127.0.0.1 8501:30001/TCP 13d

Then you need to run minikube tunnel to establish a tunnel to your service.

Now you can query your service using the external IP

$ curl -d ‘{“instances”: [1.0, 2.0, 5.0]}’ -X POST http://127.0.0.1:8501/v1/models/half_plus_two:predict
{
“predictions”: [2.5, 3.0, 4.5]
}