kubeui makes kubectl more user friendly. This is still kubectl but enhanced with fzf.

However, kubectl slows you down - requires heavy keyboard typing. In order to alleviate interaction with kubernetes API and describe the fields associated with each supported API resource directly in the Terminal, kubectl was complemented by fzf.

• fzf (command-line fuzzy finder) - https://github.com/junegunn/fzf
• yq (portable command-line YAML processor)
• Optional: tmux (terminal multiplexer)
• Optional: trivy (security scanner)
• Optional: tmuxp (tmux sessions manager)
• Optional: kubent (the test for removed or deprecated APIs. You can fine more inforamtion here: https://medium.com/@magelan09/unleashing-unlimited-potential-the-secret-to-kubecuis-unmatched-flexibility-4a41ba7003c1)

In the basic mode you use only the k alias in your terminal window. You type commands like k get pod -A, k logs and others, do what you need to do, like search for pods, deployments, view the logs and so on. Every time you need to type a new command. This mode is what kubecui was made for. This mode makes work with kubectl a little easier.

Usually we work with more than one cluster. Every time in order to switch a context we type k config use-context command. It is a tedious and error prone process (you might deploy or delete something in the wrong context). In the NORMAL mode, we work with a simple interface which has 3 windows - dev,stg,prod. To switch between windows the Ctrl+b w shortcut is used. We still need to type commands, like being in BASIC mode, but we switch the context by switching the windows. k start activates the mode. The mode is avaliable if we install optional tmux and tmuxp packages The init.sh script creates a .tmuxp folder and copies the default.yaml file there. After that, you have to edit the file( ~/.tmuxp/default.yaml) and make sure that you use actual values instead of < ...abc... > When everything is ready execute k start

Pay attention to the red rectangle. These are your clusters. * next to cluster points to the active window/cluster.Key combinations you may find helpful:

• Ctrl+b 1,2,3,n OR Ctrl+b w - switch between windows/clusters
  • 
• Ctrl+b d - exit multi-widow session
• Ctrl+b [ - edit mode. You can move cursor, scroll up/down, select and copy text.
• Ctrl+r - if you followed the installation instructions for fzf, most probably you use fzf to browse shell history. If you didn't, you should try, cause it is fun and makes your experience with kubecui more plesant.

The most efficient and the most interactive mode. Requires minimum typing. Unfortunately this mode goes against the main concept of kubecui. You don't type commands, rather switch between windows and sessions. You enter this mode by executing k start. After k start initializes the environment you will be able to switch between sessions. Each session corresponds to a single context (dev,stage or prod). The shortcut for switching between sessions - Ctrl+b s. Each session has 10 windows(for pods, deployments, logs, ingresses, configmaps, secrets, services, PV, PVC and one empty window for any commands). Why are there 10 windows? To make it easier to switch between them. Quick switching - Ctrl+b number from 0-9. Or Ctrl+b w.

• It pops up interface with windows list, use arrow keys to choose the window.

How to build and run kubecui (the darkside mode) using docker.

1. kubectl must be installed on your laptop. Because kubecui must use the same version of it, inside a docker container.
2. There are lines in the Dockerfile:

  RUN cp /root/kubecui/tmux-profiles/dev.yaml.init ~/.tmuxp/dev.yaml && sed -i 's/<KUBECONFIG_PATH>/"~\/\.kube\/dev\.yml"/g' ~/.tmuxp/dev.yaml
  RUN cp /root/kubecui/tmux-profiles/stg.yaml.init ~/.tmuxp/stg.yaml&& sed -i 's/<KUBECONFIG_PATH>/"~\/\.kube\/stg\.yml"/g' ~/.tmuxp/stg.yaml
  RUN cp /root/kubecui/tmux-profiles/prod.yaml.init ~/.tmuxp/prod.yaml&& sed -i 's/<KUBECONFIG_PATH>/"~\/\.kube\/prod\.yml"/g' ~/.tmuxp/prod.yaml

They add kubeconfig files path and name (~/.kube/dev.yml, ~/.kube/stg.yml, ~/.kube/prod.yml) to the kubecui config file (kubecui will search these particular file names). Later, you will mount kubeconfig directory from your laptop to a docker container -v <HOME_DIR_PATH>/.kube:/root/.kube, which means your's laptop /.kube/ directory must contain dev.yml, stg.yml, prod.yml. If you want to use different file names, instead of default ones, you must change them the Dockerfile sed -i 's/<KUBECONFIG_PATH>/"~\/\.kube\/BLA_BLA_SOME_NAME\.yml"/g'

3. Build the image: docker build -t kubecui:latest .
4. Run kubecui: docker run -it --rm --name kubecui -v <PATH>/kubectl:/usr/local/bin/kubectl -v <HOME_DIR_PATH>/.kube:/root/.kube kubecui:latest

IMPORTANT: For AWS EKS clusters you will additionally need -v <HOME_DIR_PATH>/.aws:/root/.aws

There is no script which leads you through the process of installation. This is done intentionally because usually tools like apt, snap, yum and so on, require root privileges, and we want the process to be transparent, at least at the earlier stage.

• fzf - Follow the installation instructions https://github.com/junegunn/fzf#installation
  • Optional: advanced kubectl command completion - https://github.com/junegunn/fzf/wiki/examples#kubectl
  • For exmaple:
    • k -n Hit [ TAB key ]
    • k get Hit [ TAB key ]
• yq - snap install yq OR apt install yq
• Optional: apt install tmux tmuxp

• git clone https://github.com/pymag09/kubecui.git

• chmod +x kubecui.sh kui_start.sh

• init.sh
• source ~/.bashrc

make sure that ~/.bashrc contains this line:

• source /<PATH>/kubecui.sh

In addition to the line for BASIC mode, make sure that ~/.bashrc contains this line too:

• export KUI_PATH="<the path to the directory where kui_start.sh is>

Also the ~/.tmuxp directory must exist and contain the default.yaml file. Remmember to update the file and replace < ...abc... > with actual values. Finally, check the kui_start.sh file and make sure that tmuxp load default line is uncommented and tmuxp load dev stg prod remains commented out

All the same as for NORMAL mode but instead of default.yaml the directory ~/.tmuxp must contain three files instead:

• dev.yaml
• stg.yaml
• prod.yaml

kui_start.sh. tmuxp load default is commented out. tmuxp load dev stg prod is uncommented

Every time you run k start it executes kui_start.sh. The script initializes the sessions. For example, if you work with AWS EKS and MFA is a part of the authentication process you can put an awsume command in the very beginning of the kui_start.sh.

I know that not all of you use bash. I really hope that you know how to do the same for other shells. At this moment I am not able to test kubecui for all the most popular shells like zsh.

Unless otherwise specified, fzf starts in "extended-search mode" where you can type in multiple search terms delimited by spaces. e.g. ^music .mp3$ sbtrkt !fire

Most probabaly you are falimiar with or already use k alias for kubectl. If so, you will be happy to hear, that kubecui is k alias in essence.

• k get <OBJECT>
• k logs
• k expain
• k explain <OBJECT>
• k get events --all-namespaces
  • Get all events. sort by first and last seen
• k config use-context
  • Singale panel. Switch between kubeconfig contexts
• k config set ns
  • Singale panel. Set default namespace.
• k start
  • starts cluster per window session
• k stop
  • stop and clear session. Without stop next time you run k start it will suggest you to return to previous session

• https://medium.com/@magelan09/unleashing-unlimited-potential-the-secret-to-kubecuis-unmatched-flexibility-4a41ba7003c1
• https://medium.com/@magelan09/kubectl-on-steroids-there-is-life-beyond-k9s-5c214e878c83
• https://medium.com/@magelan09/kubecui-enhanced-interactive-kubectl-dark-side-mode-720a1f19b0bf
• https://medium.com/@magelan09/navigating-kubernetes-effortlessly-with-kubecui-new-shortcuts-and-features-part-3-df3ab0449518
• https://medium.com/@magelan09/kubecui-ephemeral-containers-and-more-part-4-f069d7ebc405
• https://medium.com/@magelan09/kubecui-has-reached-its-first-big-milestone-part-5-in-the-series-of-the-articles-c520c3c33ab1