Can Krew be Installed as Super User (su)?

[sosukeinu]

Hello,

I'm trying to use krew with RKE2 which runs by default as the superuser. When I install krew in the userspace, kubectl (which is located at /var/lib/rancher/rke2/bin/kubectl and KUBECONFIG is here /etc/rancher/rke2/rke2.yaml) has no idea where or how to access krew when I type sudo /var/lib/rancher/rke2/bin/kubectl krew update.

I know it is probably pretty simple, but is there a trick to getting krew to work for superuser rather than current user?

Thank you for your time,
Jon

[ahmetb]

I'm trying to understand wdym by

kubectl [...] has no idea where or how to access krew

kubectl looks at $PATH to find plugins. So when you type kubectl krew it will look for kubectl-krew executable in directories listed in $PATH.

If you followed the installation instructions on Krew website, the binary is likely at $HOME/.krew/bin/kubectl-krew which won't work when you run kubectl as a different user. You may want to symlink that path to /usr/local/bin.

Krew installs any state into $HOME/.krew of the current user. You can override that with a custom path specified through $KREW_ROOT environment variable. Maybe you can set $KREW_ROOT to somewhere like /usr/local/krew and symlink /usr/local/krew/bin/kubectl-krew to /usr/local/bin/kubectl-krew so that multiple users can use the same krew installation+plugins at the same time (granted, $KREW_ROOT variable must be set in their shells as well).

/kind support

[k8s-triage-robot]

The Kubernetes project currently lacks enough contributors to adequately respond to all issues.

This bot triages un-triaged issues according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue as fresh with /remove-lifecycle stale
• Close this issue with /close
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle stale

[k8s-triage-robot]

The Kubernetes project currently lacks enough active contributors to adequately respond to all issues.

This bot triages un-triaged issues according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue as fresh with /remove-lifecycle rotten
• Close this issue with /close
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle rotten

[carlosaya]

I have been trying this as well and thus far am unable to get it to work. I am doing the following to install krew:

(
  export KREW_ROOT=/var/opt/krew
  set -x; cd "$(mktemp -d)" &&
  OS="$(uname | tr '[:upper:]' '[:lower:]')" &&
  ARCH="$(uname -m | sed -e 's/x86_64/amd64/' -e 's/\(arm\)\(64\)\?.*/\1\2/' -e 's/aarch64$/arm64/')" &&
  KREW="krew-${OS}_${ARCH}" &&
  curl -fsSLO "https://github.com/kubernetes-sigs/krew/releases/latest/download/${KREW}.tar.gz" &&
  tar zxvf "${KREW}.tar.gz" &&
  ./"${KREW}" install krew
  sudo chmod -R 0755 /var/opt/krew/store/krew/
)

I have the following in /etc/profile.d/krewpath.sh:

export KREW_ROOT=/var/opt/krew
export PATH=$PATH:/var/opt/krew/bin

When I login, I can see the following...

$ echo $KREW_ROOT
/var/opt/krew

$ echo $PATH
/home/packer/.local/bin:/home/packer/bin:/usr/local/bin:/usr/bin:/usr/local/sbin:/usr/sbin:/var/opt/krew/bin

$ ls -lah /var/opt/krew/bin
total 0
drwxr-xr-x 2 root root 26 Nov  4 17:30 .
drwxr-xr-x 6 root root 59 Nov  4 17:30 ..
lrwxrwxrwx 1 root root 36 Nov  4 17:30 kubectl-krew -> /var/opt/krew/store/krew/v0.4.4/krew

$ ls -lah /var/opt/krew/store/krew/v0.4.4/krew
-rwxr-xr-x 1 root root 12M Nov  4 17:30 /var/opt/krew/store/krew/v0.4.4/krew

Everything there looks good as far as I can tell, yet...

$ kubectl krew
Error: unknown command "krew" for "kubectl"

$ kubectl plugin list
Unable to read directory "/home/packer/.local/bin" from your PATH: open /home/packer/.local/bin: no such file or directory. Skipping...
Unable to read directory "/home/packer/bin" from your PATH: open /home/packer/bin: no such file or directory. Skipping...
: no such file or directory. Skipping...bin\r" from your PATH: open /var/opt/krew/bin
The following compatible plugins are available:

/usr/bin/kubectl-anthos
  - error: unable to identify /usr/bin/kubectl-anthos as an executable file: stat /usr/bin/kubectl-anthos: no such file or directory
error: one plugin warning was found

Invoking kubectl-krew directly works fine:

$ /var/opt/krew/bin/kubectl-krew
krew is the kubectl plugin manager.
You can invoke krew through kubectl: "kubectl krew [command]..."
<snip>

Any ideas? 🙏

[carlosaya]

Following up, it works fine if I copy the binaries to /usr/bin. Maybe kubectl doesn't follow symlinks? Had the issue with both kubectl krew and kubectl hns (which I installed via krew).