Issue writing a shell script using df

cobra4444 · 05-13-2013, 03:42 PM

Hey i'm new to writing shell scripts and I'm trying to write a script that checks how full my disks are and notifies me when they are above 90% full. The problem i'm having is that one of the Filesystem is long enough that it is moves into the next colum of output from df there by forcing the rest of the information to the next line and throughing off what information I'm pulling out of it. I was wondering if there was any way around this problem.

Below is the script I'm using:

#!/bin/sh
df -H | grep -vE '^Filesystem|tmpfs|cdrom' | awk '{ print $5 " " $1 }' | while read output;
do
echo $output
usep=$(echo $output | awk '{ print $1}' | cut -d'%' -f1 )
partition=$(echo $output | awk '{ print $2 }' )
if [ $usep -ge 90 ]; then
echo "Running out of space \"$partition ($usep%)\" on $(hostname) as on $(date)" |
mail -s "Alert: Almost out of disk space $usep%" myemailaddress@somewhere.com
fi
done

The output for df looks something like this
Filesystem Size Used Avail Use% Mounted on
/dev/sda3 20G 3.9G 15G 21% /
/dev/mapper/vg_data-lv_home
1.4T 1.3T 74G 95% /home

The last two lines on the output of df are the lines that are giving me trouble.

dayid · 05-13-2013, 04:42 PM

For clarity:

Code:

df |  # asks df for information, -h/-H is unecessary since we only are concerned with percentages anyways.
awk '{ sub (/%/, "", $5 ); # removes the % from the output so awk can compare the integer
if ( $1 ~ /^\// ) { # only is concerned with output from df beginning with /, ignores udev/tmpfs/none
    if ( $5 > 90 ) { # compares the % output from df to whatever % we'd like to be notified at
        print $6 " on " $1 " is nearly full at " $5 "% usage." # prints lines that match with some other useful info
        }
    }
}'

For simplicity:

Code:

df | awk '{ sub (/%/, "", $5 ); if ( $1 ~ /^\// ) { if ( $5 > 90 ) { print $6 " on " $1 " is nearly full at " $5 "% usage." }}}'

Where 90 is the % you want a warning at. e.g.,

Code:

dayid@clugg:/proc$ df -h | grep ^/
/dev/sda1             9.9G  4.2G  5.3G  45% /
/dev/sda6              19G  182M   18G   1% /tmp
/dev/sdd2             398G  216G  182G  55% /web
/dev/sda7              29G  1.5G   26G   6% /var
/dev/sda2              92G   78G   15G  85% /home

Using the above script, only with '80' instead of '90', will return:

Code:

/home on /dev/sda2 is oversize at 85% usage.

Does this solve the wrap issue? If not, please respond with what your $TERM is, along with which distribution you're using that is wrapping like this.

rknichols · 05-13-2013, 08:47 PM

If you use the "-P" (--portability) option, df will not insert the extra line breaks:

Code:

# df /mnt/tmp
Filesystem           1K-blocks      Used Available Use% Mounted on
/dev/mapper/bananamangosweetpealemontree
                      72248132  13498548  55079596  20% /mnt/tmp
# df -P /mnt/tmp
Filesystem         1024-blocks      Used Available Capacity Mounted on
/dev/mapper/bananamangosweetpealemontree  72248132  13498548  55079596      20% /mnt/tmp
#

cobra4444 · 05-14-2013, 07:15 AM

Thanks for the help. Just adding the -p solved the problem I was having.