Kubelet fails with an "extra space" prefixed in the label value

[surajssd]

Description

Self hosted Kubelet fails to start on the worker pools where the key value pair is ill-defined.

Environment and steps to reproduce

1. Set-up:

Create a cluster on any supported provider with following worker pool config:

  worker_pool "buggy" {
    labels = {
      "foo" = " bar"
    }
  }

Notice the space, at the start, in value of the label.

2. Task:

Once the cluster is deployed, monitor the state of kubelet pods running on above nodes by running following command:

kubectl -n kube-system get pods -o wide -l k8s-app=kubelet

3. Error:

And the kubelet on that node is in CrashLoopBackOff state. The reason being there is an extra space in the value field, logs from the kubelet pod:

/usr/local/bin/kubelet --node-ip=10.88.72.135  --anonymous-auth=false --authentication-token-webhook
--authorization-mode=Webhook --client-ca-file=/etc/kubernetes/ca.crt --cluster_dns=10.3.0.10
--cluster_domain=cluster.local --cni-conf-dir=/etc/cni/net.d --config=/etc/kubernetes/kubelet.config
--kubeconfig=/var/lib/kubelet/kubeconfig --bootstrap-kubeconfig=/etc/kubernetes/kubeconfig
--rotate-certificates --lock-file=/var/run/lock/kubelet.lock --network-plugin=cni --pod-manifest-path=/etc/kubernetes/manifests
--read-only-port=0 --volume-plugin-dir=/var/lib/kubelet/volumeplugins
--node-labels=foo= bar,node.kubernetes.io/node=metallb.lokomotive.io/my-asn=65000,metallb.lokomotive.io/peer-asn=65530
--register-with-taints=foo=bar:NoSchedule --address=10.88.72.135

[REDACTED]

--volume-stats-agg-period duration                         Specifies interval for kubelet to calculate and cache the volume disk usage for all pods and volumes.  To disable volume calculations, set to 0. (default 1m0s) (DEPRECATED: This parameter should be set via the config file specified by the Kubelet's --config flag. See https://kubernetes.io/docs/tasks/administer-cluster/kubelet-config-file/ for more information.)
F1223 07:04:11.626238   22453 server.go:163] unknown command: bar,node.kubernetes.io/node=
goroutine 1 [running]:
k8s.io/kubernetes/vendor/k8s.io/klog/v2.stacks(0xc000124001, 0xc000ada2c0, 0x5b, 0xad)

[REDACTED]

The culprit being the space in flag --node-labels: --node-labels=foo= bar,n....

Expected behavior

We should sanitise the input before passing it along.

[invidian]

I think we can return an error when there is trailing whitespace in label key or value.

[knrt10]

Also checking for taints