Question

I’m really new to UNIX/Shell Scripting I’m trying to extract disk usage from numerous servers. So what I’m trying to do is making a shell script that runs
df -g > diskusage.txt to obtain following table and extract ** data from below

   Filesystem        Size      Used    Avail    Use%        Mounted on
  /dev/ibm_lv       84.00     56.81     33%     637452     5% /usr/IBM
  /dev/apps_lv      10.00      9.95    **1%**     5        1% /usr/apps
  /dev/upi_lv      110.00     85.85   **22%**   90654      1% /usr/app/usr
  user08:/backup   2000.00   1611.22    20%    177387      1% /backup

Depending on the server, there are more file systems but i only want /usr/apps/usr,/usr/apps disk usage regardless of the number of filesystem. (/usr/apps/usr,/usr/apps will always located at last three row)

I’m pretty sure there are simpler ways than reading last 3 lines -> disregard last line -> search for % on each line.

If there is better way to extract these data, please let me know.

Answer 1

df -g | awk '/\/usr\/app/ {print $4}'

That gets you the available percentages, but it doesn’t tell you which one goes with which. You can always include the mountpoint in the output, but then you still have to do some parsing to get the numbers out, something like this:

    while read avail mount; do 
       echo "$mount has $avail available"
    done < <(df -g | awk '/\/usr\/app/ {print $4, $NF}')