Kubernetes pod in minikube can't access a service - kubernetes

Question

I can't access a service from a pod, when I run the curl serviceIP:port command from my pod console, I get the following error:
root#strongswan-deployment-7bc4c96494-qmb46:/# curl -v 10.111.107.133:80
* Trying 10.111.107.133:80...
* TCP_NODELAY set
* connect to 10.111.107.133 port 80 failed: Connection timed out
* Failed to connect to 10.111.107.133 port 80: Connection timed out
* Closing connection 0
curl: (28) Failed to connect to 10.111.107.133 port 80: Connection timed out
Here is my yaml file:
apiVersion: apps/v1
kind: Deployment
metadata:
name: strongswan-deployment
namespace: default
spec:
replicas: 1
selector:
matchLabels:
app: strongswan
template:
metadata:
labels:
app: strongswan
spec:
containers:
- name: strongswan-container
image: 192.168.39.1:5000/mystrongswan
ports:
- containerPort: 80
command: ["/bin/bash", "-c", "--"]
args: ["while true; do sleep 30; done;"]
securityContext:
privileged: true
imagePullSecrets:
- name: dockerregcred
---
apiVersion: v1
kind: Service
metadata:
namespace: default
name: strongswan-service
spec:
selector:
app: strongswan
ports:
- port: 80 # Port exposed to the cluster
protocol: TCP
targetPort: 80 # Port on which the pod listens
I tried with an Nginx pod and this time it works, I am able to connect to the Nginx service with the curl command.
I don't see where the problem comes from, since it works for the Nginx pod. What I did wrong?
I use minikube :
user#user-ThinkCentre-M91p:~/minikube$ minikube version
minikube version: v1.20.0
commit: c61663e942ec43b20e8e70839dcca52e44cd85ae
EDIT
My second pod yaml file
apiVersion: apps/v1
kind: Deployment
metadata:
name: godart-deployment
namespace: default
spec:
replicas: 1
selector:
matchLabels:
app: godart
template:
metadata:
labels:
app: godart
spec:
containers:
- name: godart-container
image: 192.168.39.1:5000/mygodart
ports:
- containerPort: 9020
imagePullSecrets:
- name: dockerregcred
---
apiVersion: v1
kind: Service
metadata:
namespace: default
name: godart-service
spec:
selector:
app: godart
ports:
- port: 9020 # Port exposed to the cluster
protocol: TCP
targetPort: 9020 # Port on which the pod listens
The error :
[root#godart-deployment-648fb8757c-6mscv /]# curl -v 10.104.206.191:9020
* About to connect() to 10.104.206.191 port 9020 (#0)
* Trying 10.104.206.191...
* Connection timed out
* Failed connect to 10.104.206.191:9020; Connection timed out
* Closing connection 0
curl: (7) Failed connect to 10.104.206.191:9020; Connection timed out
the dockerfile :
FROM centos/systemd
ENV container docker
RUN yum -y update; yum clean all
RUN yum -y install systemd; yum clean all; \
(cd /lib/systemd/system/sysinit.target.wants/; for i in *; do [ $i == systemd-tmpfiles-setup.service ] || rm -f $i; done); \
rm -f /lib/systemd/system/multi-user.target.wants/*;\
rm -f /etc/systemd/system/*.wants/*;\
rm -f /lib/systemd/system/local-fs.target.wants/*; \
rm -f /lib/systemd/system/sockets.target.wants/*udev*; \
rm -f /lib/systemd/system/sockets.target.wants/*initctl*; \
rm -f /lib/systemd/system/basic.target.wants/*;\
rm -f /lib/systemd/system/anaconda.target.wants/*;
VOLUME [ "/sys/fs/cgroup" ]
COPY /godart* /home
RUN yum install -y /home/GoDart-3.3-b10.el7.x86_64.rpm
RUN yum install -y /home/GoDartHmi-3.3-b10.el7.x86_64.rpm
CMD ["/usr/sbin/init"]
EDIT EDIT:
I solved my problem by adding a file that can respond to an http request, this is the file:
var http = require('http');
var handleRequest = function(request, response) {
console.log('Received request for URL: ' + request.url);
response.writeHead(200);
response.end('Hello World!');
};
var www = http.createServer(handleRequest);
www.listen(9020, "0.0.0.0");
To make it work you must have a Node js environment installed.
Run the script with the command:node filename.js
And after that I am able to curl my services.
I don't really understand why it works now, does anyone have an explanation ?
Thank you

Answer 1

Your strongswan-container container is using bash -c -- "while true; do sleep 30; done;" as command.
The sleep command obviously does not listen to any port.
When you try to curl your service on port 80, a TCP connection is attempted towards the Pod on port 80, but since there is no such port listening in the Pod the connection attempt fails.

Answer 2

how can I fix this error without using the sleep command?
If I good understand your question I know 2 solutions of your problem. First you can understand how work CrashLoopBackOff. Then you can change Container restart policy. The most important field should be: lastProbeTime. This means Timestamp of when the Pod condition was last probed.
Second solution should be creating a readiness probe. You can read more about it also here.