Take the 2-minute tour ×
Server Fault is a question and answer site for professional system and network administrators. It's 100% free, no registration required.

I am trying to find the biggest files in the /export/home directory and add up (sum) their sizes.

Script:

#!/bin/bash
filename=hostnames
> export_home.log

while read -r -a line
do
    hostname=${line//\"}
    echo $hostname":" >> export_home.log
    ssh -n -t -t $hostname "sudo find /export/home -type f -mtime +180 -size +250000k -exec du -hsk {} \;" >> export_home.log
done < "$filename"

Example output:

server-34:
 210M   /export/home/john/142933-02/failsafe_archive
 178M   /export/home/john/137137-09/failsafe_archive
server-35:
server-36:
 142M   /export/home/marc/bdb/db-5.2.36.tar
 446M   /export/home/marc/sunfreeware/git/git-1.7.6-sol10-x86-local
 1.4G   /export/home/marc/mysql/mysql-5.5.16-solaris10-x86_64.tar
 1.1G   /export/home/marc/mysql/mysql-5.5.16-solaris10-i386.tar
server-37:

This script does perfectly what it should do, but now how do I also get the total size of ALL files that are found based on the results in the export_home?

I am planning to make a few adjustments for this script to find the total size of log-directories and local back-up directories in order to get a better insight in combined disk-usage over multiple servers. I am not sure how I would be able to find a total disk usage.

share|improve this question
1  
First, switch your du command to always output MB. Then run the output of the find command into awk/cut and save the first field. This should be the size of the file in MB. Stash this in a variable and add to it as you loop through. Echo this out at the end. (Sorry don't have time right now to script up the code for you) –  jeffatrackaid Jun 12 '13 at 14:23
add comment

1 Answer 1

up vote 2 down vote accepted

When I use -k, I don't get "M" and "G" suffixes. If you're getting suffixes, then you should use jeffatrackaid's suggestion to make the sizes scaled to the same magnitude.

There are many ways you can total the values. Here's one:

#!/bin/bash
filename=hostnames
# no need to clear the file, just move the output redirection to the end of the loop

# are you reading into an array to split the line (hostname would be in ${line[0]})?
while read -r -a line
do
    hostname=${line//\"}
    echo "$hostname:"
    ssh -n -t -t "$hostname" "sudo find /export/home -type f -mtime +180 -size +250000k -exec du -hsk {} \;"
done < "$filename" | tee export_home.log | awk '{t += $1} END {print "grand total:", t}'

If your version of find supports it, try using + instead of \; - it's faster.

    ssh -n -t -t "$hostname" "sudo find /export/home -type f -mtime +180 -size +250000k -exec du -hsk {} +"
share|improve this answer
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.