Kubernetes initContainers to copy file and execute [duplicate] - kubernetes

Question

This question already has an answer here:
How to share a file from initContainer to base container in Kubernetes
(1 answer)
Closed 4 months ago.
The community reviewed whether to reopen this question 4 months ago and left it closed:
Original close reason(s) were not resolved
I have a situation where we will have list of IP addresses (which are coming from the config map) then we need to validate these IP Addresses (i.e. check if they are accessible from this machine) then return the first accessible IP address so that application can access this ip address to process further actions.
I got the know that we can use InitContainers to do this stuff. But my question is how can we run a shell script in the initcontainer to identify the accessible IP Address and set it in the Environmental variable so that application process this further.

Answer 1

InitContainers can communicate with other, normal containers through volumes.
You can use emptyDir volume type, which is a directory that allows the pod to store data for the duration of its life cycle.
apiVersion: v1
kind: Pod
metadata:
name: pod-name
spec:
volumes:
- name: addresses
emptyDir: {}
initContainers:
- name: ip-selector
image: your-image
volumeMounts:
- name: addresses
mountPath: /path/to/ip/addresses
containers:
- name: ip-handler
image: your-image
volumeMounts:
- name: addresses
mountPath: /path/to/ip/addresses/handler
readOnly: true # optional
Your initContainer can now save .env file with addresses in /path/to/ip/addresses path and then your normal container can read this file from /path/to/ip/addresses/handler path.

Answer 2

Option : 1
Once you get the IP inside the init container you can create the secret with value and use it.
initContainers:
- name: secret
image: gcr.io/cloud-builders/kubectl:latest
command:
- sh
- -c
- kubectl create secret mysecret ... -o yaml | kubectl apply -f -
containers:
- name: test-container
image: image-uri:latest
envFrom:
- secretRef:
name: mysecret
Option : 2
You can also use the shared mount option to write IP into the file, and inside the container, you can run the command to set the file contents to Env. So when your main container start it will start with the IP as Env.
command: [bash, -c, "source /tmp/env && service command"]
Example
apiVersion: v1
kind: Pod
metadata:
name: test-pod
spec:
containers:
- name: busybox
image: busybox
command: [bash, -c, "source /tmp/env && service command"]
volumeMounts:
- name: workdir
mountPath: /tmp/env
initContainers:
- name: ip-check
image: busybox:1.28
command:
- Command or script to check the IP write to > /tmp/env
volumeMounts:
- name: workdir
mountPath: "/tmp"
dnsPolicy: Default
volumes:
- name: workdir
emptyDir: {}