All posts by Brandon Beahm

This is a bash script I wrote for quickly grabbing the S.M.A.R.T status on all drives attached to a system along with their current temperatures. It uses ‘smartctl’ from smartmontools (www.smartmontools.org) to grab the SMART information from drives and presents snippets of that information in a nicely formatted table, including:

• path of the drive
• temperature reading
• model number
• serial number
• overall SMART status

 

Here’s what it looks like, on one of my servers

Feel free to grab this script’s source code below; using or altering as you’d like

 

#! /bin/bash
##
# prints formatted SMART results for all drives
# tested and working on: Ubuntu 18.04.1 LTS (Bionic Beaver)
##
echo "================================================================================"
echo "DRIVE::Temp::Model::Serial::Health Status" | awk -F:: '{printf "%-7s%-6s%-22s%-20s%s\n", $1, $2, $3, $4, $5}'
echo "~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~"
for i in $(lsblk | grep -E "disk" | awk '{print $1}')
do
DevSupport=`smartctl -a /dev/$i | awk '/SMART support is:/{print $0}' | awk '{print $4}' | tail -1`
if [ "$DevSupport" == "Enabled" ]
then
DevTemp=`smartctl -a /dev/$i | awk '/Temperature/{print $0}' | awk '{print $10 "C"}'`
DevSerNum=`smartctl -a /dev/$i | awk '/Serial Number:/{print $0}' | awk '{print $3}'`
DevName=`smartctl -a /dev/$i | awk '/Device Model:/{print $0}' | awk '{print $4}'`
DevStatus=`smartctl -a /dev/$i | awk '/SMART overall-health/{print $0}' | awk '{print $1" "$5" "$6}'`
echo [$i]::$DevTemp::$DevName::$DevSerNum::$DevStatus | awk -F:: '{printf "%-7s%-6s%-22s%-20s%s\n", $1, $2, $3, $4, $5}'
fi
done
##
# now find drives that don't have SMART enabled and warn user about these drives
##
echo "--------------------------------------------------------------------------------"
for i in $(lsblk | grep -E "disk" | awk '{print $1}')
do
DevSupport=`smartctl -a /dev/$i | awk '/SMART support is:/{print $0}' | awk '{print $4}' | tail -1`
if [ "$DevSupport" != "Enabled" ]
then
echo [$i]::$DevSupport | awk -F:: '{printf "%-6s **ERROR!!! SMART Support Status: %s\n", $1, $2}'
fi
done
echo "================================================================================"

I’m still new to the Linux/Unix universe and scripting on it. Please let me know if you run into any problems or if you have suggestions on how I can improve my script or feel free to give my any feedback you may have!

 

Script was built using the following:

—   Ubuntu: 18.04.1 LTS – Bionic Beaver
—   Smartmontools: 6.6 – SVN rev 4324
—   GNU bash: 4.4.19(1)-release (x86_64-pc-linux-gnu)

**Drives must have SMART support, along with enabling SMART in the system’s BIOS

 

For now this script is pretty basic barely skims the surface on managing drives properly, therefore you shouldn’t rely on this as the only form of health checks for your drives.

I plan to work more on drive monitoring scripts and hopefully will be able to offer more in the future. I’d love to see any improvements and variants you create on this scripts, or scripts you use for drive monitoring. Feel free to share them in the comments!

 

I’ve enjoyed scripting on Linux and look forward to learning more, thanks for reading

Say you have a larger text file with a bunch of lines that you don’t need or don’t want in there, how do you remove them? When working with multiple large files, such as log files, this can be a a major waste of time when done it by hand.

If you are lucky enough that the lines you need to filter out from your data have a common pattern (which just happens to be the case with most properly formatted log files) then properly learning the ‘grep’ command can save you massive amounts of time. It’s also really not that hard as you only need to learn how to utilize various options and/or special characters in conjunction with ‘grep’

 

Here’s a small snippet from a larger ‘.bash_history’ file I was working with in CentOS 7.

The .bash_history file is a log file which each user has in their home directory, this file contains all commands the user ran in the bash shell

This specific bash history file has lines that show the epoch timestamp before each command that was actually entered by the user.
This could be very useful in certain situations where you need the time that a command was entered. Yet these timestamps can be problematic if you have a script trying to access this file’s information and your script isn’t configured to read information with this very specific format.

How can we remove these unwanted lines?

Well this small snippet of only 8 lines and it would be easy enough to remove any lines by hand. The full log file I was working with had 254 lines to work with which would be considered small in terms of log files, but still quite a huge pain if you had to do it by hand!

One utility that would fit this job perfectly would be ‘grep’ which typically searches through a file line by line and spitting back lines that contain the matching pattern that you are searching for. You can also use the ‘-v’ option with grep to “invert” the information you receive back from it; so instead of grep showing you lines that DO match it will show you lines that DON’T match.

 

[simterm] $ grep -v pattern filename.txt
[/simterm] (this would output all lines from a file named ‘filename.txt’ which don’t have the word ‘pattern’)

This example would still output any lines that any variant of the word ‘pattern’ which where the word isn’t in all lower-case letters, such as ‘Pattern’ or ‘PATTERN’ or ‘pattErn’. This is because grep does case-sensitive searches by default.

 

[simterm] $ grep -v -i pattern filename.txt
[/simterm]

 

[simterm] grep -vi pattern filename.txt
[/simterm] (note: typically shouldn’t/can’t combine ‘long options’ which are those that are started with a doubledash ‘––’ )

The specific log file that I was working with didn’t require any case sensitive or insensitive searches though, as I’m simply trying remove lines that start with pound-sign.

 

[simterm] grep -v ^\# .bash_history
[/simterm]

 

… This can be quite confusing to understand if you’re just starting out.  Let’s break it down! 

 

 

Now that you know some useful grep basics
Let’s work on using the outputted data that ‘grep’ gives us in various ways, such as actually viewing this data in a better format or saving it for later by using pipes or redirects.

 

[simterm] grep -v ^\# .bash_history | more
[/simterm] a simple pipe, by using using ‘ | ‘ between commands

 

[simterm] grep -v ^\# .bash_history > .bash_history_cleaned
[/simterm] a simple redirect, using ‘ > ‘ and then the new file’s name

 

[simterm] grep -v ^\# .bash_history > temp.file && cat temp.file > .bash_history && rm temp.file
[/simterm] This example redirects the cleaned data into temporary file, then outputs that temporary file back into the original file. Overwriting all of the old file’s data with the cleaned up data!

 

What exactly is the ‘&&’ from that last command?
&&  is an ‘operator’ and is used as a way to string multiple commands. Specifically with the ‘&&’ operator, the next command is dependent on the command before it. This means the next command won’t execute unless the command before it ran successfully.

(more on ‘operands’ https://www.gnu.org/software/bash/manual/bashref.html#Lists)

 

 

— Hope this has helped you on your learning journey, thanks for reading! —