Empty an s3 bucket of all object versions and delete markers in batches of 400

empty_bucket.sh

#!/usr/bin/env bash

set -eEo pipefail
shopt -s inherit_errexit >/dev/null 2>&1 || true

if [[ ! "$#" -eq 2 || "$1" != --bucket ]]; then
    echo -e "USAGE: $(basename "$0") --bucket <bucket>"
    exit 2
fi

# $@ := bucket_name
empty_bucket() {
    local -r bucket="${1:?}"
    for object_type in Versions DeleteMarkers; do
        local opt=() next_token=""
        while [[ "$next_token" != null ]]; do
            page="$(aws s3api list-object-versions --bucket "$bucket" --output json --max-items 400 "${opt[@]}" \
                        --query="[{Objects: ${object_type}[].{Key:Key, VersionId:VersionId}}, NextToken]")"
            objects="$(jq -r '.[0]' <<<"$page")"
            next_token="$(jq -r '.[1]' <<<"$page")"
            case "$(jq -r .Objects <<<"$objects")" in
                '[]'|null) break;;
                *) opt=(--starting-token "$next_token")
                   aws s3api delete-objects --bucket "$bucket" --delete "$objects";;
            esac
        done
    done
}

empty_bucket "${2#s3://}"

@zishanj I only ever needed to delete them all, to be able to delete the bucket. But you could probably achieve that just by modifying the page query on line 18.

@bitlifter thanks for the hint, I lowered the number in the script.

When you have files with very long name, it will fail with Argument list too long, even with a very low bunch of --max-items

So I rewrite a bit your function. Thanks anyway, it helps ! so I share mine. It's very slow as it loop over each file, but at least it works

function empty_bucket() {
  for object_type in Versions DeleteMarkers; do
    aws s3api list-object-versions \
      --bucket "${BUCKET_NAME}" \
      --output json \
      --query="[{Objects: ${object_type}[].{Key:Key, VersionId:VersionId}}]" > /tmp/${object_type}.json
    is_null=$(jq -c '.[0].Objects' /tmp/${object_type}.json)
    if [ "${is_null}" = "null" ]; then
      echo "Nothing to clean-up for ${object_type}"
    else
      jq -r -c '.[].Objects[] | .VersionId, .Key' /tmp/${object_type}.json | while read -r version && read -r key; do
        aws s3api delete-object \
          --bucket "${BUCKET_NAME}" \
          --key "${key}" \
          --version-id "${version}" \
          --no-paginate \
          --no-cli-pager \
          --output text
      done
    fi
  done
}