worker node Scale Down Problem In kubernetes

[ahmadamirahmadi007]

problem

i can not scale down my worker node in the kubernetes cluster. when I check the cloudstack log I found the following error:

2025-01-22 07:53:20,644 ERROR [c.c.u.s.SshHelper] (API-Job-Executor-125:ctx-6dc068a9 job-18918 ctx-e219b42e) (logid:e7efb240) SSH execution of command sudo /opt/bin/kubectl drain test2-node-1948cfb170b --ignore-daemonsets --delete-local-data has an error status code in return. Result output: error: unknown flag: --delete-local-data
2025-01-22 07:53:20,645 WARN [c.c.k.c.a.KubernetesClusterActionWorker] (API-Job-Executor-125:ctx-6dc068a9 job-18918 ctx-e219b42e) (logid:e7efb240) Draining node: test2-node-1948cfb170b on VM : test2-node-1948cfb170b in Kubernetes cluster : test2 unsuccessful

versions

CloudStack 4.19.1.3
kubernetes version 1.31.1
kubectl verion v1.31.1

The steps to reproduce the bug

form Scale Kubernetes cluster in cluster part menu i scale down worker node from 2 to 1

What to do about it?

No response

[weizhouapache]

--delete-local-data has been deprecated in kubernetes

let me create a pr for it

[ahmadamirahmadi007]

So why don't I encounter this problem when I use kubernetes version 1.29.9, if I use kubectl 1.29.9, it is easily scaled down.

[DaanHoogland]

as @weizhouapache said, there is an option that is deprecated/replaced in version 1.31.1; --delete-local-data ==> --delete-emptydir-data

[weizhouapache]

as @weizhouapache said, there is an option that is deprecated/replaced in version 1.31.1; --delete-local-data ==> --delete-emptydir-data

yes, it has been deprecated for many versions.

finally it is dropped in 1.31:

https://github.com/kubernetes/kubernetes/blob/a78983906f4cf6f94029a3e6cf3171ef42b395d7/CHANGELOG/CHANGELOG-1.31.md?plain=1#L1338

[DaanHoogland]

fixed by #10234