I was doing the CKA and stumbled upon a question I couldn't figure out. I don't remember all the details, but it goes something like this:
Get the top 1 node/pod by CPU consumption and place it in a file at {path}.
kubectl top nodes/pod --sort-by cpu <-- this orders by ascending. So you have to hardcode the last node/pod.If you need to extract out the name of the top pod and save it in the file you can do this:
Let us say you have 3 pods:
$ kubectl top pod --sort-by cpu NAME CPU(cores) MEMORY(bytes) nats-depl-6c4b4dfb7c-tjpgv 2m 4Mi project-depl-595bbd56db-lb6vb 8m 180Mi auth-depl-64cccc484f-dn7w5 4m 203Mi
You can do this:
$ kubectl top pod --sort-by cpu | head -2 | tail -1 | awk {'print $1'}
chat-depl-6dd798699b-l7wcl #### < == returns name of the pod
## you can redirect it to any file
$ kubectl top pod --sort-by cpu | head -2 | tail -1 | awk {'print $1'} > ./file_name
$ cat ./file_name
chat-depl-6dd798699b-l7wcl #### < == returns name of the pod