Error when re-using edge devices

[robert-sanfeliu]

I tried to make a deployment with an edge device that wasn't fresh. It had already been used to deploy a cluster (successfully).

The deployment failed on the worker initialization with this error:

357t3@ip-192-168-55-2.ec2.internal;worker-tstedgamd20241212351-tstedgamd2024121235_Task_install_0;10:47:55] Saving to: ‘install-kube-u22-wg.sh’
[357t3@ip-192-168-55-2.ec2.internal;worker-tstedgamd20241212351-tstedgamd2024121235_Task_install_0;10:47:55] 0K .... 100% 76.7M=0s
[357t3@ip-192-168-55-2.ec2.internal;worker-tstedgamd20241212351-tstedgamd2024121235_Task_install_0;10:47:55] 2024-09-12 10:47:55 (76.7 MB/s) - ‘install-kube-u22-wg.sh’ saved [4459/4459]
[357t3@ip-192-168-55-2.ec2.internal;worker-tstedgamd20241212351-tstedgamd2024121235_Task_install_0;10:47:55] INFO [Thu Sep 12 10:47:55 UTC 2024]: Configuration started!
[357t3@ip-192-168-55-2.ec2.internal;worker-tstedgamd20241212351-tstedgamd2024121235_Task_install_0;10:47:55] INFO [Thu Sep 12 10:47:55 UTC 2024]: Logs are saved at: /var/log/kube-install.25296.log
[357t3@ip-192-168-55-2.ec2.internal;worker-tstedgamd20241212351-tstedgamd2024121235_Task_install_0;10:47:55] INFO [Thu Sep 12 10:47:55 UTC 2024]: Updating the package list.
[357t3@ip-192-168-55-2.ec2.internal;worker-tstedgamd20241212351-tstedgamd2024121235_Task_install_0;10:48:31] INFO [Thu Sep 12 10:48:31 UTC 2024]: Checking for apt lock
[357t3@ip-192-168-55-2.ec2.internal;worker-tstedgamd20241212351-tstedgamd2024121235_Task_install_0;10:48:31] INFO [Thu Sep 12 10:48:31 UTC 2024]: Exited the while loop, time spent: 0
[357t3@ip-192-168-55-2.ec2.internal;worker-tstedgamd20241212351-tstedgamd2024121235_Task_install_0;10:48:31] INFO [Thu Sep 12 10:48:31 UTC 2024]: Waiting for lock task ended properly.
[357t3@ip-192-168-55-2.ec2.internal;worker-tstedgamd20241212351-tstedgamd2024121235_Task_install_0;10:48:31] INFO [Thu Sep 12 10:48:31 UTC 2024]: Installing curl
[357t3@ip-192-168-55-2.ec2.internal;worker-tstedgamd20241212351-tstedgamd2024121235_Task_install_0;10:48:31] INFO [Thu Sep 12 10:48:31 UTC 2024]: Installing Docker
[357t3@ip-192-168-55-2.ec2.internal;worker-tstedgamd20241212351-tstedgamd2024121235_Task_install_0;10:48:32] INFO [Thu Sep 12 10:48:32 UTC 2024]: Adding Kubernetes Repo
[357t3@ip-192-168-55-2.ec2.internal;worker-tstedgamd20241212351-tstedgamd2024121235_Task_install_0;10:48:32] INFO [Thu Sep 12 10:48:32 UTC 2024]: Checking for apt lock
[357t3@ip-192-168-55-2.ec2.internal;worker-tstedgamd20241212351-tstedgamd2024121235_Task_install_0;10:48:42] INFO [Thu Sep 12 10:48:42 UTC 2024]: Exited the while loop, time spent: 10
[357t3@ip-192-168-55-2.ec2.internal;worker-tstedgamd20241212351-tstedgamd2024121235_Task_install_0;10:48:42] INFO [Thu Sep 12 10:48:42 UTC 2024]: Waiting for lock task ended properly.
[357t3@ip-192-168-55-2.ec2.internal;worker-tstedgamd20241212351-tstedgamd2024121235_Task_install_0;10:48:42] ERROR [Thu Sep 12 10:48:42 UTC 2024]: Kubernetes repo can't be added!

This error comes from the execution of the script install-kube-u22-wg.sh (https://raw.githubusercontent.com/eu-nebulous/sal-scripts/main/k8s/install-kube-u22-wg.sh) on line

curl -fsSL https://pkgs.k8s.io/core:/stable:/v1.26/deb/Release.key | sudo gpg --dearmor -o /etc/apt/keyrings/kubernetes-apt-keyring.gpg || { log_print ERROR "Kubernetes repo can't be added!"; exit $EXITCODE; }
sudo apt-get update

If I SSH to the edge node and I try to run the script. I get:

ubuntu@worker-tstedgamd20241212351-tstedgamd2024121235:~$ curl -fsSL https://pkgs.k8s.io/core:/stable:/v1.26/deb/Release.key | sudo gpg --dearmor -o /etc/apt/keyrings/kubernetes-apt-keyring.gpg
File '/etc/apt/keyrings/kubernetes-apt-keyring.gpg' exists. Overwrite? (y/N)

My assumption is that installation failed because '/etc/apt/keyrings/kubernetes-apt-keyring.gpg' already existed.