Unix/Linux tutorials for beginners
Безопасный ввод данных в скриптах BASH
I needed to create a very simple bash script that required a password to be provided using keyboard and then used further in the script. Turns out, there's just a way to do it with the bash built-in function called read.
[Read more…] about Secure Keyboard Input in bash Scripts
Now and then I have to find some information in text files – logs, config files, etc. with a unique enough approach: I have a hostname or parameter name to search by, but the actual information I'm interested in is not found in the same line. Simple grep will just not do – so I'm using one of the really useful command line options for the grep command.
By default, grep shows you just the lines of text files that match the parameter you specify.
I'm looking for the IP address of the home office server called "server" in .ssh/config file.
Here's what default grep shows me:
greys@maverick:~ $ grep server .ssh/config Host server
So it's useful in a sense that this confirms that I indeed have a server profile defined in my .ssh/config file, but it doesn't show any more lines of the config so I can't really see the IP address of my "server".
This is where the -A option (A – After) for grep becomes really useful: specify number of lines to show in addition to the line that matches search pattern and enjoy the result!
I'm asking grep to show me 2 more lines after the "server" line:
greys@maverick:~ $ grep -A2 server .ssh/config
Host server
HostName 192.168.1.55
ForwardAgent yes
Super simple but very powerful way of using grep. I hope you like it!
Back to basics day: a very simple tip on working with one of the most useful things available in Unix and Linux filesystems: symbolic links (symlinks).
When using long-form ls output, symlinks will be shown like this:
greys@mcfly:~/proj/unixtutorial $ ls -la
total 435736
-rwxr-xr-x 1 greys staff 819 14 Dec 2018 .gitignore
drwxr-xr-x 3 greys staff 96 20 Nov 09:27 github
drwxr-xr-x 4 greys staff 128 20 Nov 10:58 scripts
lrwxr-xr-x 1 greys staff 30 10 Dec 20:40 try -> /Users/greys/proj/unixtutorial
drwxr-xr-x 44 greys staff 1408 20 Nov 10:54 wpengine
Show Symlink Destination with readlink
As with pretty much everything in Unix, there's a simple command (readlink) that reads a symbolic link and shows you the destination it's pointing to. Very handy for scripting:
greys@mcfly:~/proj/unixtutorial $ readlink try /Users/greys/proj/unixtutorial
It has a very basic syntax: you just specify a file or directory name, and if it's a symlink you'll get the full path to the destination as the result.
If readlink returns nothing, this means the object you're inspecting isn't a symlink at all. Based on the outputs above, if I check readlink for the regular scripts directory, I won't get anything back:
greys@mcfly:~/proj/unixtutorial $ readlink scripts
greys@mcfly:~/proj/unixtutorial $
I'm slowly improving my Python skills, mostly by Googling and combining multiple answers to code a solution to my systems administration tasks. Today I decided to write a simpe converter that takes Epoch Time as a parameter and returns you the time and date it corresponds to.
Most of the functionality is done using the fromtimestamp function of the datetime module. But because I also like seeing time in UTC, I decided to use timezone module as well.
#!/Library/Frameworks/Python.framework/Versions/3.6/bin/python3
import sys
from datetime import datetime, timezone
if len(sys.argv)>1:
print ("This is the Epoch time: ", sys.argv[1])
try:
timestamp = int(sys.argv[1])
if timestamp>0:
timedate = datetime.fromtimestamp(timestamp)
timedate_utc = datetime.fromtimestamp(timestamp, timezone.utc)
print ("Time/date: ", format(timedate))
print ("Time/date in UTC: ", format(timedate_utc))
except ValueError:
print ("Timestamp should be a positive integer, please.")
else:
print ("Usage: epoch.py <EPOCH-TIMESTAMP>")
FIXME: I'll revisit this to re-publish script directly from GitHub.
Here's how you can use the script:
greys@maverick:~/proj/python $ ./epoch.py 1566672346 This is the Epoch time: 1566672346 Time/date: 2019-08-24 19:45:46 Time/date in UTC: 2019-08-24 18:45:46+00:00
I implemented basic checks:
Do you see anything that should be changed or can be improved? Let me know!
Removing files and directories is a very common task, so in some environments support engineers or automation scripts delete hundreds of files per day. That's why I think it's important to be familiar with different ways and safety mechanisms you should use when it comes to removing Unix directories. This article has a number of principles that should help you make your day-to-day files and directories operations safer.
[Read more…] about Advice: Safely Removing a File or Directory in Unix
Last night I finished a fun mini project as part of Unix Tutorials Projects. I have writted a basic enough script that can be added as root cronjob for automatically controlling keyboard backlight on my Dell XPS 9380.
As I've written just a couple of days ago, it's actually quite easy to turn keyboard backlight on or off on a Dell XPS in Linux (and this probably works with other Dell laptops).
Armed with that knowledge, I've written the following script:
#!/bin/bash
WORKDIR=/home/greys/scripts/backlight
LOCKFILE=backlight.kbd
LOGFILE=${WORKDIR}/backlight.log
KBDBACKLIGHT=`cat /sys/devices/platform/dell-laptop/leds/dell::kbd_backlight/brightness`
HOUR=`date +%H`
echo "---------------->" | tee -a $LOGFILE
date | tee -a $LOGFILE
if [ $HOUR -lt 4 -o $HOUR -gt 21 ]; then
echo "HOUR $HOUR is rather late! Must turn on backlight" | tee -a $LOGFILE
BACKLIGHT=3
else
echo "HOUR $HOUR is not too late, must turn off the backlight" | tee -a $LOGFILE
BACKLIGHT=0
fi
if [ $KBDBACKLIGHT -ne $BACKLIGHT ]; then
echo "Current backlight $KBDBACKLIGHT is different from desired backlight $BACKLIGHT" | tee -a $LOGFILE
FILE=`find ${WORKDIR} -mmin -1440 -name ${LOCKFILE}`
echo "FILE: -$FILE-"
if [ -z "$FILE" ]; then
echo "No lock file! Updating keyboard backlight" | tee -a $LOGFILE
echo $BACKLIGHT > /sys/devices/platform/dell-laptop/leds/dell::kbd_backlight/brightness
touch ${WORKDIR}/${LOCKFILE}
else
echo "Lockfile $FILE found, skipping action..." | tee -a $LOGFILE
fi
else
echo "Current backlight $KBDBACKLIGHT is the same as desired... No action needed" | tee -a $LOGFILE
fi
This is what my script does when you run it as root (it won't work if you run as regular user):
Log file looks like this:
greys@xps:~/scripts $ tail backlight/backlight.log ----------------> Tue May 28 00:10:00 BST 2019 HOUR 00 is rather late! Must turn on backlight Current backlight 2 is different from desired backlight 3 Lockfile /home/greys/scripts/backlight/backlight.kbd found, skipping action... ----------------> Tue May 28 00:15:00 BST 2019 HOUR 00 is rather late! Must turn on backlight Current backlight 2 is different from desired backlight 3 Lockfile /home/greys/scripts/backlight/backlight.kbd found, skipping action...
I have added this script to the root user's cronjob. In Ubuntu 19.04 running on my XPS laptop, this is how it was done:
greys@xps:~/scripts $ sudo crontab -e [sudo] password for greys:
I then added the following line:
*/5 * * * * /home/greys/scripts/backlight.sh
Depending on where you place similar script, you'll need to update full path to it from /home/greys/scripts. And then update WORKDIR variable in the script itself.
Here are just a few things I plan to improve:
I'm decorating my tmux setup and needed to confirm colour numbers for some elements of the interface. Turns out, it's simple enough to show all the possible colours with a 1-liner in your favourite Unix shell – bash shell in my case.
I'll explain how this works in full detail sometime in a separate post, but for now will just give you an example and show how it works:
So, in this example, this is how we achieve colorized text output:
Here's how we get the colours now: we create a loop from 1 until 255 (0 will be black) and then use the ESC syntax changing colour to $COLOR variable value. We then output the $COLOR value which will be a number:
for COLOR in {1..255}; do echo -en "\e[38;5;${COLOR}m${COLOR} "; done; echo;
Here's how running this will look in a propertly configured 256-color terminal:
Here's the same 1-liner converted into proper script for better portability and readability:
#!/bin/bash
for COLOR in {1..255}; do
echo -en "\e[38;5;${COLOR}m"
echo -n "${COLOR} "
done
echo
If you save this as bash-256-colours.sh and chmod a+rx bash-256-colours.sh, you can now run it every time you want to refresh your memory or pick different colours for some use.
If you're using CentOS, Fedora or Red Hat Linux, you are probably familiar with the yum package manager. One of the really useful options for yum is checking whether there are any available updates to be installed.
If you use check-update parameter with yum, it will show you the list of any available updates:
root@centos:~ # yum check-update Loaded plugins: fastestmirror, langpacks Loading mirror speeds from cached hostfile * base: rep-centos-fr.upress.io * epel: mirror.in2p3.fr * extras: rep-centos-fr.upress.io * updates: ftp.pasteur.fr ansible.noarch 2.7.8-1.el7 epel datadog-agent.x86_64 1:6.10.1-1 datadog libgudev1.x86_64 219-62.el7_6.5 updates nginx.x86_64 1:1.15.9-1.el7_4.ngx nginx oci-systemd-hook.x86_64 1:0.1.18-3.git8787307.el7_6 extras polkit.x86_64 0.112-18.el7_6.1 updates systemd.x86_64 219-62.el7_6.5 updates systemd-libs.i686 219-62.el7_6.5 updates systemd-libs.x86_64 219-62.el7_6.5 updates systemd-python.x86_64 219-62.el7_6.5 updates systemd-sysv.x86_64 219-62.el7_6.5 updates
One thing that I didn't know and am very happy to discover is that yum check-update is actually meant for shell scripting. It returns a specific code after running, you can use the value to decide what do to next.
As usual: return value 0 means everything is fully updated, so no updates are available (and no action is needed). A value of 100 would mean you have updates available.
All we need to do is check the return value variable $? for its value in something like this:
#!/bin/bash
yum check-update
if [ $? == 100 ]; then
echo "You've got updates available!"
else
echo "Great stuff! No updates pending..."
fi
Here is how running this script would look if we saved the script as check-yum-updates.sh script:
root@s2:~ # ./check-yum-updates.sh Loaded plugins: fastestmirror, langpacks Loading mirror speeds from cached hostfile * base: rep-centos-fr.upress.io * epel: mirror.in2p3.fr * extras: rep-centos-fr.upress.io * updates: ftp.pasteur.fr ansible.noarch 2.7.8-1.el7 epel datadog-agent.x86_64 1:6.10.1-1 datadog libgudev1.x86_64 219-62.el7_6.5 updates nginx.x86_64 1:1.15.9-1.el7_4.ngx nginx oci-systemd-hook.x86_64 1:0.1.18-3.git8787307.el7_6 extras polkit.x86_64 0.112-18.el7_6.1 updates systemd.x86_64 219-62.el7_6.5 updates systemd-libs.i686 219-62.el7_6.5 updates systemd-libs.x86_64 219-62.el7_6.5 updates systemd-python.x86_64 219-62.el7_6.5 updates systemd-sysv.x86_64 219-62.el7_6.5 updates You've got updates available!
I'll revisit this post soon to show you a few more things that can be done with yum check-update functionality.
pwd command, and you probably know, reports the current working directory in your Unix shell. PWD stands for Present Working Directory. In addition to pwd, there's also a special variable – one of the user environment variables – called PWD, that you can also use.
Just to remind you, pwd command is a super simple way of confirming where you are in the filesystem tree.
Usually, your shell session starts in your home directory. For me and my username greys, this means /home/greys in most distributions:
greys@xps:~$ pwd /home/greys
If I then use cd command to visit some other directory, pwd command will help me confirm it:
greys@xps:~$ cd /tmp greys@xps:/tmp$ pwd /tmp
Most Unix shells have PWD as a variable. It reports the same information as pwd command but saves children processes the trouble of using pwd command and getpwd system call just to confirm the working directory of their parent process.
So, you can just do this to confirm $PWD value:
greys@xps:/tmp$ echo $PWD /tmp
… which really helps in shell scripting, cause you can do something like this:
#!/bin/bash echo "Home directory: $HOME" echo "Current directory: $PWD" if [ $HOME != $PWD ]; then echo "You MUST run this from your home directory!" exit else echo "Thank you for running this script from your home directory." fi
When we run this, the script will compare standard $HOME variable (your user's homedir) to the $PWD variable and will behave differently if they match.
I've created and saved pwd.sh in my projects directory for bash scripts: /home/greys/proj/bash:
greys@xps:~/proj/bash$ ./pwd.sh Home directory: /home/greys Current directory: /home/greys/proj/bash You MUST run this from your home directory!
If I now change back to my home directory:
greys@xps:~/proj/bash$ cd /home/greys/
… the script will thank me for it:
greys@xps:~$ proj/bash/pwd.sh Home directory: /home/greys Current directory: /home/greys Thank you for running this script from your home directory.
Have fun using pwd command and $PWD variable in your work and shell scripting!
Part of managing any Unix system is keeping an eye on the vital log files.
Today I was discussing one of such scenarios with a friend and we arrived at a pretty cool example involving awk command and eventually a bash command substitution.
Let's say we have a directory with a bunch of log files, all constantly updated at different times and intervals. Here's how I may get the last 10 lines of the output from the most recent log file:
root@vps1:/var/log# cd /var/log root@vps1:/var/log# ls -altr *log -rw-r--r-- 1 root root 32224 Jul 10 22:49 faillog -rw-r----- 1 syslog adm 0 Jul 25 06:25 kern.log -rw-r--r-- 1 root root 0 Aug 1 06:25 alternatives.log -rw-r--r-- 1 root root 2234 Aug 8 06:34 dpkg.log -rw-rw-r-- 1 root utmp 294044 Aug 15 22:32 lastlog -rw-r----- 1 syslog adm 12248 Aug 15 22:35 syslog -rw-r----- 1 syslog adm 5160757 Aug 15 22:40 auth.log
Ok, now we just need to get that filename from the last line (auth.log).
Most obvious way would be to use tail command to extract the last line, and awk to show the 9th parameter in that line – which would be the filename:
root@vps1:/var/log# ls -altr *log | tail -1 | awk '{print $9}'
auth.log
Pretty cool, but can be optimised using awk's END clause:
root@vps1:/var/log# ls -altr *log | awk 'END {print $9}'
auth.log
Alright. Now we wanted to show the 10 lines of output, which we can use tail -10 for.
A really basic approach is to assing the result of the line we're using to a variable in Bash, and then access that variable, like this:
root@vps1:/var/log# FILE=`ls -altr *log | tail -1 | awk '{print $9}'`
root@vps1:/var/log# tail -10 ${FILE}
Aug 15 22:40:37 vps1 sshd[26578]: pam_unix(sshd:auth): authentication failure; logname= uid=0 euid=0 tty=ssh ruser= rhost=159.65.145.196
Aug 15 22:40:39 vps1 sshd[26578]: Failed password for invalid user Fred from 159.65.145.196 port 47934 ssh2
Aug 15 22:40:39 vps1 sshd[26578]: Received disconnect from 159.65.145.196 port 47934:11: Normal Shutdown, Thank you for playing [preauth]
Aug 15 22:40:39 vps1 sshd[26578]: Disconnected from 159.65.145.196 port 47934 [preauth]
Aug 15 22:41:15 vps1 sshd[26580]: Connection closed by 51.15.4.190 port 44958 [preauth]
Aug 15 22:42:02 vps1 sshd[26585]: Connection closed by 13.232.227.143 port 40054 [preauth]
Aug 15 22:43:23 vps1 sshd[26587]: Connection closed by 51.15.4.190 port 52454 [preauth]
Aug 15 22:44:08 vps1 sshd[26589]: Connection closed by 13.232.227.143 port 47542 [preauth]
Aug 15 22:45:01 vps1 CRON[26604]: pam_unix(cron:session): session opened for user root by (uid=0)
Aug 15 22:45:01 vps1 CRON[26604]: pam_unix(cron:session): session closed for user root
But an ever shorter (better?) way to do this would be to use the command substitution in bash: the output of a command becomes the command itself (or string value in our case).
Check it out:
root@vps1:/var/log# tail -10 $(ls -altr *log | tail -1 | awk '{print $9}')
Aug 15 22:42:02 vps1 sshd[26585]: Connection closed by 13.232.227.143 port 40054 [preauth]
Aug 15 22:43:23 vps1 sshd[26587]: Connection closed by 51.15.4.190 port 52454 [preauth]
Aug 15 22:44:08 vps1 sshd[26589]: Connection closed by 13.232.227.143 port 47542 [preauth]
Aug 15 22:45:01 vps1 CRON[26604]: pam_unix(cron:session): session opened for user root by (uid=0)
Aug 15 22:45:01 vps1 CRON[26604]: pam_unix(cron:session): session closed for user root
Aug 15 22:45:26 vps1 sshd[26610]: Connection closed by 51.15.4.190 port 59872 [preauth]
Aug 15 22:46:15 vps1 sshd[26612]: Connection closed by 13.232.227.143 port 55030 [preauth]
Aug 15 22:46:23 vps1 sshd[26608]: Connection closed by 18.217.190.140 port 40804 [preauth]
Aug 15 22:47:28 vps1 sshd[26614]: Connection closed by 51.15.4.190 port 39044 [preauth]
Aug 15 22:48:20 vps1 sshd[26616]: Connection closed by 13.232.227.143 port 34286 [preauth]
So in this example $(ls -altr *log | tail -1 | awk '{print $9}') is a substituion – bash executes the command in the parenthesis and then passes the resulting value to further processing (becoming a parameter for the tail -10 command).
In our command above, we're essentially executing the following command right now:
root@vps1:/var/log# tail -10 auth.log
only auth.log is always the filename of the log file that was updated the latest, so it could become syslog or dpkg.log if they're updated before next auth.log update.