In zsh, how to store this fragment as an environment variable to be outputted in another command

Question

I'd like to run the following:

kubectl get po -o custom-columns=NAME:'{.metadata.name}',OWNER:'{.metadata.ownerReferences[0].name}',OWNER_KIND:'{.metadata.ownerReferences[0].kind}'

I'd like to store I store the part after po in a variable so that I could run something like:

kubectl get po $K8OWNER

I have tried adding to my .zshrc

export K8SO="-o custom-columns=NAME:'{.metadata.name}',OWNER:'{.metadata.ownerReferences[0].name}',OWNER_KIND:'{.metadata.ownerReferences[0].kind}'"

but this doesn't work. How could I do this?

Answer 1

hope you are enjoying your kubernetes journey !

I got a workaround for you:

You can create your own plugin: https://kubernetes.io/docs/reference/kubectl/#examples-creating-and-using-plugins

Lets do it together, its as simple as saying hello:

First create a file named "kubectl-podowner", with:

#!/bin/sh

# this plugin prints the pods with the custom columns header of their owners
kubectl get po -o custom-columns=NAME:".metadata.name",OWNER:".metadata.ownerReferences[0].name",OWNER_KIND:".metadata.ownerReferences[0].kind"

now, let's give the x rights to this file:

chmod a+x ./kubectl-podowner

and move it to a location in our path:

 sudo mv ./kubectl-podowner /usr/local/bin
 sudo chown root:root /usr/local/bin

now lets see if this worked:

? kubectl podowner
NAME                            OWNER                     OWNER_KIND
nginx-deploy-6bdc4445fd-5qlhg   nginx-deploy-6bdc4445fd   ReplicaSet
nginx-deploy-6bdc4445fd-pgkhb   nginx-deploy-6bdc4445fd   ReplicaSet
nginx-deploy-6bdc4445fd-xdz59   nginx-deploy-6bdc4445fd   ReplicaSet

PERFECT FEATURE ! (I love it)

You can do the same with kubectl-rsowner:

#!/bin/sh

# this plugin prints the replicasets with the custom columns header of their owners
kubectl get rs -o custom-columns=NAME:'{.metadata.name}',OWNER:'{.metadata.ownerReferences[0].name}',OWNER_KIND:'{.metadata.ownerReferences[0].kind}'

It works also perfectly when fixing ^M errors:

? kubectl rsowner
NAME                      OWNER          OWNER_KIND
nginx-deploy-6bdc4445fd   nginx-deploy   Deployment

PS: here is some troubleshooting tips i faced:

If when executing kubectl podowner you get this error:

? kubectl podowner
Error: no such file or directory

This is probably because your kubectl-podowner file contains ^Ms like this:

#!/bin/sh^M
^M
# this plugin prints the pods with the custom columns header of their owners^M
kubectl get po -o custom-columns=NAME:".metadata.name",OWNER:".metadata.ownerReferences[0].name",OWNER_KIND:".metadata.ownerReferences[0].kind"^M

Just vi the kubectl-podowner file and follow this ./configure : /bin/sh^M : bad interpreter