disk_usage_hotspots.sh

#!/usr/bin/env bash

set -euo pipefail

# disk_usage_hotspots.sh
# Lists largest directories/files under a path to identify disk hotspots.

usage() {
  cat <<'EOF'
Usage: ./disk_usage_hotspots.sh [options]

Options:
  --path DIR             Root path to scan (default: /)
  --depth N              Directory depth for du scan (default: 2)
  --top N                Number of entries to show (default: 30)
  --files                Include largest files listing
  -h, --help             Show help
EOF
}

SCAN_PATH="/"
DEPTH=2
TOP_N=30
SHOW_FILES=false

while [[ $# -gt 0 ]]; do
  case "$1" in
    --path) SCAN_PATH="$2"; shift 2 ;;
    --depth) DEPTH="$2"; shift 2 ;;
    --top) TOP_N="$2"; shift 2 ;;
    --files) SHOW_FILES=true; shift ;;
    -h|--help) usage; exit 0 ;;
    *) echo "Unknown option: $1"; usage; exit 1 ;;
  esac
done

if [[ ! -d "$SCAN_PATH" ]]; then
  echo "Path not found: $SCAN_PATH"
  exit 1
fi

echo "Top $TOP_N directories under $SCAN_PATH (depth=$DEPTH):"
du -x -h --max-depth="$DEPTH" "$SCAN_PATH" 2>/dev/null | sort -hr | head -n "$TOP_N"

if [[ "$SHOW_FILES" == true ]]; then
  echo
  echo "Top $TOP_N files under $SCAN_PATH:"
  find "$SCAN_PATH" -xdev -type f -printf "%s %p\n" 2>/dev/null | sort -nr | head -n "$TOP_N" | awk '{printf "%.2f MB %s\n", $1/1024/1024, substr($0, index($0,$2))}'
fi