duboisf

emulate -RL zsh

setopt err_return warn_create_global
local API_RESOURCE="$1"
local ctx
local kind

if [[ -z "$API_RESOURCE" ]]; then
    print "Usage: kubectl_search_api_resources <API_RESOURCE>" >&2
    print "Will search for the api resources that match the regex API_RESOURCE in all clusters and all namespaces" >&2

duboisf

kubectl get no -o json
| from json
| get items
| select status.images status.nodeInfo.architecture 
| rename images arch 
| flatten
| update images.names { |in|
    if ($in | length) > 1 {
        $in | filter { $in !~ @sha }
    } else {

duboisf

In nushell, you can generate a changelog on the fly with the following:

gh release list --json name
| from json
| get name
| par-each --keep-order { |name|
    gh release view $name --json name,body,author
    | from json
 | $"# ($in.name) by ($in.author.login)\n\n($in.body)"

duboisf

Troubleshooting Redis StatefulSet: Clearing Data from a PVC When Pods Are Crashlooping

The Problem

I encountered an issue with a Redis StatefulSet in our staging environment where the pods were stuck in a crashloop. The root cause was a full Persistent Volume Claim (PVC) that had run out of space. Since this was cache data in a staging environment, I needed to clear the /data folder to resolve the issue.

The challenge was that I couldn't exec into the pod directly because the Redis container was continuously crashlooping, making standard troubleshooting approaches impossible.

The Solution

duboisf

aws ecr describe-repositories | from json | get repositories.repositoryName
| sort
| each { |repo|
    let scanOnPush = (aws ecr batch-get-repository-scanning-configuration --repository-names $repo
        | from json
        | get scanningConfigurations
        | first
        | get scanOnPush
    )
    print $"Repository: ($repo), Scanning on push: ($scanOnPush)"

duboisf

-- Example of using coroutines with slow IO operations in Neovim.
-- It shows how to run coroutines sequentially and in "parallel".
-- The word "parallel" is in quotes because Lua coroutines are not actually
-- running in parallel, but the IO operations that they start are.
-- The coroutines are paused and resumed as the IO operations complete.
-- To run this example, put this file somewhere and do `:luafile path/to/coroutines.lua` in nvim.

local co = coroutine

vim.cmd "new"

duboisf

Get all s3 bucket policies in an aws account using nushell

In parallel! Saves output to a file so you can simply do open /tmp/bucket-policies.nuon after:

^aws s3api list-buckets
| from json
| get Buckets.Name
| par-each { |bucket|
 let policy = try {

duboisf

fleetctl user create --name 'Gitops User' \
  --password "$(cat /dev/urandom | tr -cd '[:graph:]' | head -c 32)" \
  --email [email protected] \
  --api-only \
  --global-role admin

duboisf

Multi-architecture docker builds on Linux

Install binfmt if ls /proc/sys/fs/binfmt_misc/qemu* returns nothing:

docker run --privileged --rm tonistiigi/binfmt --install all

Verify that it worked, you should see stuff by running ls /proc/sys/fs/binfmt_misc/qemu*

duboisf

#!/bin/bash

# This script diffs the content of two container images
# USAGE: ./diff-container-images.sh <registry>/<repo> <tag1> <tag2>

# strict mode
set -euo pipefail

registry_repo=$1
tag1=$2