Skip to content

USAGE.md

Latest commit

 

History

History
111 lines (92 loc) · 3.78 KB

USAGE.md

File metadata and controls

111 lines (92 loc) · 3.78 KB

Usage

ketall demo

If you installed via krew do

kubectl get-all

Options

  • --only-scope=cluster will only show cluster level resources, such as ClusterRole, Namespace, or PersistentVolume.
  • --only-scope=namespace will only show namespaced resources, such as ServiceAccount, Role, ConfigMap, or Endpoint.
  • --selector (-l) will filter by label query, supports =, ==, and !=.(e.g. -l key1=value1,key2=value2)
  • --exclude will filter out the given resources. Accepts either resource names (e.g. componentstatuses or short form cs) or API Kinds (e.g. ComponentStatus). Defaults to [Event, PodMetrics] because those are rarely useful.
  • ...and many standard kubectl options. Have a look at kubectl get-all --help for a full list of supported flags.
  • --use-cache will consider the http cache to determine the server resources to look at. Disabled by default.
  • --allow-incomplete will show partial results when fetching the list of API resources fails. Enabled by default.
  • -v set the log level (one of debug, info, warn, error, fatal, panic).

Hint: If you do not have access to all resources, bulk fetching needs to be disabled. You can speed things up by explicitly excluding all resources which you may not access.

Examples

Get all resources...

  • ... excluding events (this is hardly ever useful)

    kubectl get-all

  • ... including events

    kubectl get-all --exclude=

  • ... created in the last minute

    kubectl get-all --since 1m

    This flag understands typical human-readable durations such as 1m or 1y1d1h1m1s.

  • ... in the default namespace

    kubectl get-all --namespace=default

  • ... at cluster level

    kubectl get-all --only-scope=cluster

  • ... using list of cached server resources

    kubectl get-all --use-cache

    Note that this may fail to show really everything, if the http cache is stale.

  • ... and combine with common kubectl options

    KUBECONFIG=otherconfig kubectl get-all -o name --context some --namespace kube-system --selector run=skaffold

Getting help

kubectl get-all help

Note that in the help, the tool is referred to as ketall, which is the standard name when installed as stand-alone tool.

Completion

Completion does currently not work when used as a kubectl plugin. When used stand-alone, you can do

source <(ketall completion bash) # for bash users
source <(ketall completion zsh)  # for zsh users

Also see ketall completion --help for further instructions.

Configuration

The command will look for the configuration file ketall (no extension) in . or $HOME/.kube/, unless overridden by the --config option. The following settings can be configured:

only-scope: cluster
namespace: default
use-cache: true
since: 1m
selector: run=skaffold,tail!=true
# only plural form or abbreviations
exclude:
- componentstatuses
- cm   # configmaps

Installation

Via krew

If you do not have krew installed, visit https://github.com/kubernetes-sigs/krew.

kubectl krew install get-all

As kubectl plugin

Most users will have installed ketall via krew, so the plugin is already correctly installed. Otherwise, rename ketall to kubectl-get_all and put it in some directory from your $PATH variable. Then you can invoke the plugin via kubectl get-all

Standalone

Put the ketall binary in some directory from your $PATH variable. For example

sudo mv -i ketall /usr/bin/ketall

Then you can invoke the plugin via ketall