Accessing the deployed service using Helm chart in Kubernetes cluster

12
votes

Currently, I am trying to deploy my microservice end point Docker image on a Kubernetes cluster by creating the Helm chart. For this, I created the chart and changed the parameters in values.yaml and deployment.yaml for port change. And also I want to access from my Angular front end. So I added service type= NodePort. And when I described the service, it gave me the port 30983 to access.

And I accessed like http://node-ip:30983/endpoint

But I am only getting the site can't be reached the message. Let me add the details of what I did here:

My values.yaml file containing the following to mention the service type:

enter image description here

And my templates/service.yaml file containing like the following:

enter image description here

And my templates/deployment.yaml file containing the following:

enter image description here

And I tried to access like the following:

http://192.168.16.177:30983/

And only getting site can't be reached.

NB: when I tried to describe the service, then I am getting the following:

enter image description here

The output of kubectl get pod --show-labels like the following image screenshot

enter image description here

Updated

And when we using kubectl describe pod command, getting like the following:

enter image description here

Updated Error

Readiness probe failed: HTTP probe failed with statuscode: 404
Liveness probe failed: HTTP probe failed with statuscode: 404

How can I access my endpoint from deployment?

2
kuberneteskubernetes-helm
please provide output of "kubectl get pod --show-labels" and "kubectl get svc" – Vasili Angapov
@VasilyAngapov - Yes sir, updated in question. And thank you for your response. – Jacob
your pod is Crashlooping. Your app is crashing for some reason. You need to read logs "kubectl logs -p POD_NAME" where POD_NAME is the name of your failing pod. – Vasili Angapov
@VasilyAngapov - I updated my result sir. Can you please look on updated question ? – Jacob
you didn't provide the logs of crashlooping pods. Without logs nobody can help you. Read my previous comment again pls. – Vasili Angapov

2 Answers

4
votes

Try this for healthcheck probes:

livenessProbe:
  tcpSocket:
    port: 8085
readinessProbe:
  tcpSocket:
    port: 8085
1
votes

try the following command docker ps -a and find the container associated with the pod. The container name should be pretty much same as the pod name with some prefix/suffix.

then look at the logs using docker logs <container_id>. Maybe that will give you clues to what it is restarting