kubectl-fzf provides a fast and powerful fzf autocompletion for kubectl.
- Seamless integration with kubectl autocompletion
- Fast completion
- Label autocompletion
- Automatic namespace switch
- go (minimum version 1.19)
- awk
- fzf
# Completion binary called during autocompletion
go install github.com/bonnefoa/kubectl-fzf/v3/cmd/kubectl-fzf-completion@main
# If you want to run the kubectl-fzf server locally
go install github.com/bonnefoa/kubectl-fzf/v3/cmd/kubectl-fzf-server@mainkubectl-fzf-completion needs to be in you $PATH so make sure that your $GOPATH bin is included:
PATH=$PATH:$GOPATH/bin
Source the autocompletion functions:
# bash version
wget https://raw.githubusercontent.com/bonnefoa/kubectl-fzf/main/shell/kubectl_fzf.bash -O ~/.kubectl_fzf.bash
echo "source <(kubectl completion bash)" >> ~/.bashrc
echo "source ~/.kubectl_fzf.bash" >> ~/.bashrc
# zsh version
wget https://raw.githubusercontent.com/bonnefoa/kubectl-fzf/main/shell/kubectl_fzf.plugin.zsh -O ~/.kubectl_fzf.plugin.zsh
echo "source <(kubectl completion zsh)" >> ~/.zshrc
echo "source ~/.kubectl_fzf.plugin.zsh" >> ~/.zshrc
You can use antigen to load it as a zsh plugin
antigen bundle robbyrussell/oh-my-zsh plugins/docker
antigen bundle bonnefoa/kubectl-fzf@main shell/You can deploy kubectl-fzf-server as a pod in your cluster.
From the k8s directory:
helm template --namespace myns --set image.kubectl_fzf_server.tag=v3 --set toleration=aToleration . | kubectl apply -f -You can check the latest image version here.
You can install kubectl-fzf-server as a systemd unit server.
# Create user systemd config
mkdir -p ~/.config/systemd/user
wget https://raw.githubusercontent.com/bonnefoa/kubectl-fzf/main/systemd/kubectl_fzf_server.service -O ~/.config/systemd/user/kubectl_fzf_server.service
# Set fullpath of kubectl-fzf-server
sed -i "s#INSTALL_PATH#$GOPATH/bin#" ~/.config/systemd/user/kubectl_fzf_server.service
# Reload to pick up new service
systemctl --user daemon-reload
# Start the server
systemctl --user start kubectl_fzf_server.service
# Automatically enable it at startup
systemctl --user enable kubectl_fzf_server.service
# Get log
journalctl --user-unit=kubectl_fzf_server.service
flowchart TB
subgraph TargetCluster
k8s[api-server]
end
subgraph Laptop
shell[Shell]
fileNode([/tmp/kubectl_fzf_cache/TargetCluster/pods])
comp[kubectl-fzf-completion]
server[kubectl-fzf-server]
end
shell -- kubectl get pods TAB --> comp -- Read content and feed it to fzf --> fileNode
server -- Write autocompletion informations --> fileNode
k8s <-- Watch --o server
kubectl-fzf-server will watch cluster resources and keep the current state of the cluster in local files.
By default, files are written in /tmp/kubectl_fzf_cache (defined by KUBECTL_FZF_CACHE)
Advantages:
- Minimal setup needed.
- Local cache is maintained up to date.
Drawbacks:
- It can be CPU and memory intensive on big clusters.
- It also can be bandwidth intensive. The most expensive is the initial listing at startup and on error/disconnection. Big namespace can increase the probability of errors during initial listing.
- It can generate load on the kube-api servers if multiple user are running it.
To create cache files necessary for kubectl_fzf, just run in a tmux or a screen
kubectl-fzf-serverIt will watch the cluster in the current context. If you switch context, kubectl-fzf-server will detect and start watching the new cluster.
The initial resource listing can be long on big clusters and autocompletion might need 30s+.
connect: connection refused or similar messages are expected if there's network issues/interruptions and kubectl-fzf-server will automatically reconnect.
flowchart TB
subgraph TargetCluster
k8s[api-server]
server[kubectl-fzf-server]
end
subgraph Laptop
shell[Shell]
comp[kubectl-fzf-completion]
end
shell -- kubectl get pods TAB --> comp
comp -- Through port forward\nGET /k8s/resources/pods --> server
k8s <-- Watch --o server
If the pod is deployed in your cluster, the autocompletion will be fetched automatically fetched using port forward.
Advantages:
- No need to run a local
kubectl-fzf-server - Only a single instance of
kubectl-fzf-serverper cluster is needed, lowering the load on thekube-apiservers.
Drawbacks:
- Resources need to be fetched remotely, this can increased the completion time. A local cache is maintained to lower this.
Once kubectl-fzf-server is running, you will be able to use kubectl_fzf by calling the kubectl completion
# Get fzf completion on pods on all namespaces
kubectl get pod <TAB>
# Open fzf autocompletion on all available label
kubectl get pod -l <TAB>
# Open fzf autocompletion on all available field-selector. Usually much faster to list all pods running on an host compared to kubectl describe node.
kubectl get pod --field-selector <TAB>
# This will fallback to the normal kubectl completion (if sourced)
kubectl <TAB>By default, the local port used for the port-forward is 8080. You can override it through an environment variable:
KUBECTL_FZF_PORT_FORWARD_LOCAL_PORT=8081
Build and test a completion with debug logs:
go build ./cmd/kubectl-fzf-completion && KUBECTL_FZF_LOG_LEVEL=debug ./kubectl-fzf-completion k8s_completion 'get pods '
Force Tab completion to use the completion binary in the current directory:
export KUBECTL_FZF_COMPLETION_BIN=./kubectl-fzf-completion
To debug Tab completion, you can activate the shell debug logs:
export KUBECTL_FZF_COMP_DEBUG_FILE=/tmp/debug
Check that the completion function is correctly sourced:
type kubectl_fzf_completion
kubectl_fzf_completion is a shell function from /home/bonnefoa/.antigen/bundles/kubectl-fzf-main/shell/kubectl_fzf.plugin.zsh
Use zsh completion debug:
kubectl get pods <C-X>?
Trace output left in /tmp/zsh497886kubectl1 (up-history to view)
To launch kubectl-fzf-server with debug logs
kubectl-fzf-server --log-level debug