disk usage summary per user

diskstats.sh

#!/bin/bash

# Find disk usage per base directory, per user
# http://unix.stackexchange.com/a/77509/28675
# http://stackoverflow.com/a/32533111/1243763

BASEDIR="$1"
MYTIME="$(date +%d%b%y_%H%M%S%Z)"

if [ -z "$BASEDIR" ] || [ ! -d "$BASEDIR" ]; then
	echo -e "Missing or incorrect argument. Specified directory "$1" is not present relative to current working directory, $(pwd)"
	exit 1
else
	echo -e "\n##### DISK USAGE REPORT - "$MYTIME" ####\n"

	find "$BASEDIR" -type f -printf '%u %k\n' | awk '{ 
                                            arr[$1] += $2 
                                        } END { 
                                            for ( i in arr ) { 
                                                print i": "arr[i]"K" 
                                            }
                                        }'
fi

## How to ##

# name this script as diskstats.sh and keep it in ~/bin/diskstats.sh

### for a specific directory, use following ###
# diskstats.sh <dir_path>
# diskstats.sh /verhaak-data/users6

### to recurse over level directories ###
# e.g., users1 to users10 disks under /verhaak-data
# cd /verhaak-data && find . -mindepth 1 -maxdepth 1 -type d -print0 | xargs -0 -L1 sh -c 'echo -e "\n#### DISK REPORT for "$0" ####\n" && ~/bin/diskstats.sh "$0"' 2>&1 | tee -a ~/verhaak-data_usage_$(whoami).log

# If you need usage report per level 1 directories within each of 10 verhaak-disks, then issue same command as follows:
# cd /verhaak-data && find . -mindepth 2 -maxdepth 2 -type d -print0 | xargs -0 -L1 sh -c 'echo -e "\n#### DISK REPORT for "$0" ####\n" && ~/bin/diskstats.sh "$0"' 2>&1 | tee -a ~/verhaak-data_usage_$(whoami).log

## END ##

I tend to enable disk quotas instead, leaving them set to 0, only using repquota to quickly report in usage.

That's very useful and I should ask my server admin to enable disk quota. Thanks!