Collection of tips on Linux




Collection of tips on Linux (mostly Debian/Ubuntu) helpful to me

Bash Keyboard Shortcuts

Esc + tSwap the two words before the cursor
Ctrl + rSearch command history
Ctrl + gCancel command history search without running command
Ctrl + lClear terminal screen
Ctrl + xList possible filename completions
Ctrl + cCancels the running command
Ctrl + zSuspends the running command
Ctrl + uDeletes entire line before the cursor
Ctrl + kDeletes entire line after the cursor
Ctrl + tSwap the two characters before the cursor
Ctrl + dClose the terminal
Ctrl + fGo forward one character
Ctrl + bGo back one character
Ctrl + aGo to the beginning of the line
Ctrl + eGo to the end of the line
Ctrl + wDelete the word before the cursor
Ctrl + yRetrieves the last word deleted or cut
Ctrl + xxToggle between current cursor position and start or end of line
Alt + uCapitalize all letters in word after cursor
Alt + lLower case all letters in word after cursor
Alt + .Use the last word of the last command


Basic configuration for new Git repository

# Set user name and e-mail address (required to do 'commit')
git config [--global] ""
git config [--global] "User Name"

# Store/cache password
git config [--global] credential.helper store
git pull

# Set the remote repository (for existing code)
git remote add origin

Add existing user to existing group

sudo usermod -a -G groupnames username

-a – append groups to group user belongs to (instead of overwrite).
groupnames – a comma-separated (no spaces!) list of group names to add user to.
User must log out and back in for group membership updates to be applied.

Swap the Caps Lock and (Right) Control keys on keyboard

The level of configurability in Linux is simply amazing. With the venerable xmodmap utility, keyboard remapping is a snap. Just add these lines to your $HOME/.xmodmap file to swap the Caps Lock and (Right) Control keys.

remove Lock = Caps_Lock
remove Control = Control_R
keysym Control_R = Caps_Lock
keysym Caps_Lock = Control_R
add Lock = Caps_Lock
add Control = Control_R

Enter Unicode characters with keyboard in Linux

In most applications in Linux, including at the command line, to enter a Unicode character, hold down <LeftCtrl> and <Shift> plus u and enter the 2- or 4-character hexadecimal Unicode code. When you release <LeftCtrl> and <Shift>, the character will be displayed. For example, to enter superscript 2 (²), which is Unicode 00B2, type <LeftCtrl>+<Shift>+u+B2; for the trademark symbol (™), which is Unicode 2122, type <LeftCtrl>+<Shift>+u+2122. Note that you can use the numeric keys across the top of the keyboard or on the numeric keypad (with NumLock enabled).

Install packages required to build application from source in Ubuntu/Debian

If you want to build an application from source for a new version of an application that has a Ubuntu/Debian package, you can use the build-dep utility to install the required dependencies in one go.

sudo apt-get build-dep PKG_NAME

where PKG_NAME is the package name, such as vim-common.

“Safe” alternative to bypassing password prompt for sudo

To avoid getting prompted for password when running commands with sudo, one common option is to append NOPASSWD:ALL to your user name in the /etc/sudoers file. Obviously, this is a security risk. Instead, you can run the sudo command with the -s (“session”) flag to allow the sudo session to be persistent until your close the terminal (end the session). To explicitly end the session run sudo -k (“kill”).

Change default editor for visudo

By default, Linux systems use the $VISUAL or $EDITOR environment variables (usually defined in your ~/.bashrc file or /etc/profile) as the default editor the visudo command. If you’d prefer to use a different editor, such as nano, you can use either of these methods.

  1. To temporarily use a different editor, run:
$ sudo EDITOR=/path/to/editor visudo

For example, to use nano, you would run:

$ sudo EDITOR=nano visudo
  1. To permanently change the default editor, edit the /etc/sudoers file (you can use the temporary method above!) and add the following line to the file near the top, but after Defaults env_reset:
Defaults    editor=/path/to/editor


Upgrade Ubuntu to non-LTS version via command line

By default, most installations of Ubuntu are configured to upgrade only to LTS (long-term support) distribution releases, which come out every two years (e.g., 18.04, 20.04, etc.). If you want to upgrade your Ubuntu installation to a non-LTS release (e.g., from Bionic Beaver [18.04] to Eoan Ermine [19.10]) you can do so via command line. Here’s how.

Update current release to latest patches

sudo apt-get install update-manager-core -y
sudo apt-get update && sudo apt-get upgrade -y && sudo apt-get dist-upgrade -y

Update upgrade manager to normal (non-LTS) setting

sudo sed -i 's/Prompt=lts/Prompt=normal/g' /etc/update-manager/release-upgrades

Change distribution reference to desired version codename* and disable third-party repositories (PPAs)

sudo sed -i 's/bionic/eoan/g' /etc/apt/sources.list
sudo sed -i 's/^/#/' /etc/apt/sources.list.d/*.list

Replace bionic and eoan above with the current and desired distribution version codenames, respectively, as appropriate. *See here for list of Ubuntu distribution codenames with associated version numbers.

Run the upgrade, remove unneeded packages, and reboot to complete update

echo '* libraries/restart-without-asking boolean true' | sudo debconf-set-selections
sudo apt-get update && sudo apt-get upgrade -y && sudo apt-get dist-upgrade -y -o Dpkg::Options::=--force-confdef -o Dpkg::Options::=--force-confnew
sudo apt-get autoremove -f -y && sudo apt-get clean -y
sudo shutdown -r now

See this article for details about forcing use of new/package maintainer’s version of configuration files. For additional details refer to this article.

Confirm new release version

lsb_release -a

Re-enable third-party repositories (PPAs) and change them to the new version codename

sudo sed -i '/deb/s/^#//g' /etc/apt/sources.list.d/*.list
sudo sed -i 's/bionic/eoan/g' /etc/apt/sources.list.d/*.list
sudo apt-get update && sudo apt-get upgrade -y

If you get any errors that a repository can’t be found (e.g., The repository ' eoan Release' does not have a Release file.), then you will need to revert these individual repositories to the earlier distribution version codename in /etc/apt/sources.list.d directory.


Bash script to toggle touchpad on and off

If you use an external mouse with your laptop, you probably want to disable your touchpad when the mouse is plugged in. Here’s how to create a simple Bash script to toggle the touchpad on and off.

At the Bash prompt, run this command to list all of your input devices, such as the keyboard, mouse, and touchpad:

$ xinput
⎡ Virtual core pointer                      	id=2	[master pointer  (3)]
⎜   ↳ Virtual core XTEST pointer              	id=4	[slave  pointer  (2)]
⎜   ↳ USB Optical Mouse                       	id=10	[slave  pointer  (2)]
⎜   ↳ **SynPS/2 Synaptics TouchPad              	id=12**	[slave  pointer  (2)]

In this example, the touchpad, has device ID 12. Next, we check the status (enabled or disabled) for this device:

$ xinput -list-props 12 | grep "Device Enabled"
	Device Enabled (**116**):	**1**

Here, the 1 means that the touchpad is enabled. Create a script named with the following contents and replace 12 and 116 with the appropriate values for your machine:

if xinput list-props **12** | grep "Device Enabled (**116**):.*1" >/dev/null
  xinput disable **12**
  notify-send -u low -i mouse "Touchpad disabled"
  xinput enable **12**
  notify-send -u low -i mouse "Touchpad enabled"

Copy to a directory in your $PATH and make it executable (chmod +x Simply run it anytime that you want to toggle the touchpad on or off.


Disable GPG checking for third-party repositories (PPAs)

When using third-party repositories (PPAs), you typically need to install GPG key. If you have trouble with GPG keys, you can configure the repository, in /etc/apt/sources.list or the custom configuration file in /etc/apt/sources.list.d/ by adding trusted=yes or allow-insecure=yes. The difference between them is that allow-insecure=yes will prompt you before allowing you to install, but trusted=yes won’t.

For example, here’s the setting used with the MongoDB repository:

deb [ arch=amd64 `allow-insecure=yes ] bionic/mongodb-org/4.0 multiverse


Reset ‘root’ password for MySQL Server (version 8.0+)

If you install MySQL Server 8.0 or later in Ubuntu without specifying the root password, you can set (reset) it as follows.

Run mysql utility with root account without password:

sudo mysql

(Note: This only works if no root password is set.)

At the MySQL prompt, use the ALTER USER command to set the desired password. It is important to specify the authentication plugin as mysql_native_password to allow applications such as phpMyAdmin to connect (see also here).

ALTER USER 'root'@'localhost' IDENTIFIED WITH mysql_native_password BY 'new_root_password';

Restart MySQL service:

sudo systemctl restart mysql

You should now be able to log in to MySQL from command prompt with the new password:

mysql -u root -pnew_root_password


Mount AWS S3 Bucket Using s3fs

s3fs is a FUSE (File System in Userspace) extension which allows you to mount an Amazon Web Services (AWS) S3 bucket as native local file system. In other words, no specialized tools are required.

We will use s3fs package from the Ubuntu repositories. You can install s3fs by building it from source; see s3fs Github repository for details.

Switch to the root user before performing the other steps:

sudo su -

Install s3fs:

apt-get install -y s3fs

Create the system s3fs password file using the appropriate AWS S3 credentials (access key ID and secret access key).

chmod 600 /etc/passwd-s3fs

/etc/passwd-s3fs can contain multiple sets of credentials (access key ID and secret access key pair combinations) with each on its own line in the file.

Create file system directories to mount S3 bucket and for caching S3 bucket contents. The cache directory is optional, but will improve performance when using S3 bucket with large number of files.

mkdir /tmp/cache
mkdir /mnt/s3-bucket
chmod 777 /tmp/cache /mnt/s3-bucket

Mount the S3 bucket using s3fs. (Note: This mount is temporary/non-persistent. See below for mounting the file system on boot using /etc/fstab.)

s3fs s3-bucket-name /mnt/s3-bucket -o passwd_file=/etc/passwd-s3fs -o allow_other,use_cache=/tmp/cache

Replace s3-bucket-name with the desired S3 bucket for the credentials specified in /etc/passwd-s3fs from above. Note that rw means to mount the file system as “read-write” (the default setting); if you want to mount as “read-only”, change this to ro.

Test the S3 bucket file system mount. You should see a “standard” file system listing. And, of course, you can use GUI file managers by navigating to /mnt/s3-bucket.

ls -lrt /mnt/s3-bucket

To mount the S3 bucket as your (non-root) user ID, at a regular (non-root) command prompt run id ${USER}. You should see something like:

id ${USER}
uid=1000(tim) gid=1000(tim) groups=1000(tim),4(adm),24(cdrom),27(sudo),30(dip),33(www-data),46(plugdev),107(input),121(lpadmin),131(lxd),133(sambashare),998(docker)

Use the uid and gid values above to run s3fs:

s3fs s3-bucket-name /mnt/s3-bucket -o passwd_file=/etc/passwd-s3fs -o allow_other,use_cache=/tmp/cache,uid=1000,umask=077,gid=1000

If you get an error about not being allowed to use allow_other as regular user, you will need to uncomment the user_allow_other line in /etc/fuse.conf FUSE configuration file.

To configure your system to automatically (“permanently”) mount the S3 bucket when it boots, do the following. (This assumes that you are still logged in as root user.)

echo s3fs#s3-bucket-name /mnt/s3-bucket fuse _netdev,rw,nosuid,nodev,allow_other,nonempty,uid=1000,umask=077,uid=1000 0 0 >> /etc/fstab

Mount (re-mount) the file system to ensure that it works properly.

mount -a

That’s it! Now you can transparently work with your S3 buckets just like they are local files.


Fix HP Pavilion Laptop lockup/freeze problem on idle/inactivity in Linux

Some HP Pavilion laptops experience problems with freezing after inactivity timeouts or other idle conditions. Some or all of the following items can help prevent such problems.

  • Disable power saving features.
  • Disable screensaver and screen lock features.
  • Add kernel boot parameters to GRUB boot menu options.
    • noapic – Disable APIC (Advanced Programmable Interrupt Controller) support.
    • idle=nomwait – Disable “mwait” for CPU idle state.


Display time and time zone details for system

For a quick check of the time and time zone details of your Linux system, run the timedatectl command. It will show you the current local time, UTC time, RTC (real-time clock from BIOS) and the time zone, along with some additional details. The command can also be used to change these settings. See the timedatectl man page for more information.

Show processor architecture information and use in shell script

Many times in a shell script, you may need to differentiate between whether your Linux platform is 32-bit or 64-bit and which processor type/architecture is used. Here are a few commands you can use to get such information.

CommandFlagsActionExample Output
uname-mDisplay machine hardware namex86_64 or i686
archAlias for uname -mx86_64
dpkg--print-architectureDisplay machine/platform architecture (Debian/Ubuntu)amd64arm64i386
dpkg-architecture--query DEB_BUILD_**ARCH**_CPUDisplay machine/platform architecture (Debian/Ubuntu)amd64arm64i386
dpkg-architecture--query DEB_BUILD_**GNU**_CPUDisplay GNU architecturex86x86_64
nprocDisplay number of CPU cores4
getconfLONG_BITDisplays 32 or 64, depending on address bus64
lscpuDetailed information about CPUN/A
lshw-C CPUSummary information about CPUN/A


Tidy up your Linux command line history with HISTIGNORE

If you use the Linux command line often, one of the greatest features is the history of commands run. Simply press the up and down arrows to navigate backward and forward through the commands or hit Ctrl+R to search. However, if you navigate around a lot and list directory contents, your history can filled with extra commands that you aren’t likely to want to from history, since they are simple enough to just run again. To prevent history from adding these to your command history, just add the HISTIGNORE variable to your .bashrc with a list of the commands to ignore separated with colons (:). Here’s an example:

HISTIGNORE="ls:ls -lrt:[bf]g:history*:exit:*shutdown*:*reboot*:[ \t]*"

In this example, we ignore ls by itself and ls -lrt, the bg and fg commands, exit, and anything starting with history. Likewise, you can see that we’ve included “dangerous” commands like shutdown and reboot that we probably don’t want to accidentally run when quickly scrolling through a long history list. And, finally, [ \t]* means to ignore any command that you enter that starts with a Space or Tab so that you can selectively run a command and have it ignored in the history.

Configure remote access on Ubuntu with XRDP

Historically, remote access on Linux was handled through the VNC (Virtual Network Computing) platform, such as with x11vnc. However, recently, the Remote Desktop Protocol (RDP) developed for Microsoft Windows has gained popularity on Linux using the XRDP tool. Here’s how to set it up on Ubuntu.

Configure remote access on Ubuntu with XRDP

Historically, remote access on Linux was handled through the VNC (Virtual Network Computing) platform, such as with x11vnc. However, recently, the Remote Desktop Protocol (RDP) developed for Microsoft Windows has gained popularity on Linux using the XRDP tool. Here’s how to set it up on Ubuntu.

  • Install XRDP from the Ubuntu repositories.
$ sudo apt-get install -y xrdp dbus-x11
  • After installation, Ubuntu will automatically launch the XRDP service. To confirm that it launched properly, run:
$ sudo systemctl status xrdp

You should see output similar to the following. The main thing to confirm is that it shows that the service is active (running) next to Active.

● xrdp.service - xrdp daemon
  Loaded: loaded (/lib/systemd/system/xrdp.service; enabled; vendor prese>
  Active: active (running) since Thu 2021-07-22 06:37:19 CDT; 29min ago
    Docs: man:xrdp(8)
 Process: 58107 ExecStartPre=/bin/sh /usr/share/xrdp/socksetup (code=exit>
 Process: 58115 ExecStart=/usr/sbin/xrdp $XRDP_OPTIONS (code=exited, stat>
Main PID: 58116 (xrdp)
   Tasks: 2 (limit: 8740)
  Memory: 16.6M
  CGroup: /system.slice/xrdp.service
          ├─58116 /usr/sbin/xrdp
          └─58124 /usr/sbin/xrdp
  • XRDP uses the /etc/ssl/private/ssl-cert-snakeoil.key SSL certificate file for authentication. Access to this file is limited to members of the ssl-cert group, so we must add the xrdp user to that group with this command:
$ sudo adduser xrdp ssl-cert
  • To ensure that XRDP service is enabled, so that it starts on system boot, run:
$ sudo systemctl enable xrdp

As in step #2 above, you can check the status again to ensure that it’s running and enabled.

  • To avoid an authentication warning on the Ubuntu machine after logging in (see step #5 below), we must add a Policy Kit rule, as indicated below.
$ sudo nano /etc/polkit-1/localauthority.conf.d/02-allow-colord.conf

polkit.addRule(function(action, subject) {
    if (( == "org.freedesktop.color-manager.create-device" || == "org.freedesktop.color-manager.create-profile" || == "org.freedesktop.color-manager.delete-device" || == "org.freedesktop.color-manager.delete-profile" || == "org.freedesktop.color-manager.modify-device" || == "org.freedesktop.color-manager.modify-profile") && subject.isInGroup("{group}"))
        return polkit.Result.YES;

Restart the xrdp service for the change to take effect.

$ sudo systemctl restart xrdp
  1. Connect to your Ubuntu machine from another machine. If you are connecting from a Windows machine, you can use the standard Remote Desktop client (Start –> Windows Accessories –> Remote Desktop Connection or Start –> Run –> mstsc.exe). Enter the IP address or hostname of your Ubuntu machine to connect. In a few seconds, you’ll be prompted to log in; log in using the same Ubuntu username and password that you usually use. (You may also be prompted about SSL certificate validation, which you will need to accept.)For connecting from another Ubuntu (or other Linux) machine, there are various RDP clients that you can use, including:
    • Remmina*
    • Vinagre*
    • TigerVNC* – To use TigerVNC with RDP, enter the IP address or hostname in the viewer connection window followed by :3389. This means to connect on TCP port 3389, which is the port that RDP listens on.
    • KRDC*
    • X2Go*
      *Indicates that these are available in standard Ubuntu repositories (i.e., install using sudo apt-get install).
  2. If you get a black (blank) screen on the remote machine after logging in remotely, you’ll need to make a simple adjustment on the Ubuntu machine (the one that you are accessing remotely) in the script to launches the X11 window manager for XRDP. Do the following:
sudo nano /etc/xrdp/

Add the lines below right below/after if block concerning /etc/profile near the bottom of the file.


After making the change, the file should looking something like this:

if test -r /etc/profile; then
        . /etc/profile


test -x /etc/X11/Xsession && exec /etc/X11/Xsession
exec /bin/sh /etc/X11/Xsession
  • Finally, restart and enable the xrdp service to ensure the change takes effect.


Prevent laptop from suspending when closing lid in Linux

If you have an old laptop that you typically access remotely (i.e., via VNC or RDP), then usually you will want to close the lid, but not have the device suspend, so that you can still connect remotely. Some Linux desktop environments (DE) have GUI options to do this, but many don’t and, even those that do, use a variety of different techniques (e.g., power management, login, etc.). Here’s the most reliable way to do it.

  • Edit the logind service configuration file as root.
$ sudo nano /etc/systemd/logind.conf
  • Locate the lines under [Login] section that start with HandleLidSwitch and uncomment them (remove the # at the start of the line, if any) and set the value to either lock (preferred) or ignore. (The current values, as commented out, are the defaults.) The lock setting will turn off the display and lock the machine, which will require you to enter your password when you open the lid, while ignore does nothing when you close the lid.
  • Save the file and reboot the machine for the changes to take effect. (You can actually just restart the logind service [i.e., sudo systemctl restart logind]. However, this will have the effect of logging you out.]

Change screen resolution in Linux in running in Windows HyperV VM

When running Linux in a Windows HyperV VM, typically, the Display configuration in the Linux instance will not have any provision to change the screen resolution. To change the resolution, you can adjust it via a command-line parameter in Grub. Edit the /etc/default/grub file as root user and append video=hyperv_fb:1152x864 to the GRUB_CMDLINE_LINUX_DEFAULT and GRUB_CMDLINE_LINUX settings. You can choose whatever resolution you prefer, such as 1024×768, 1900×1200, etc. For example:

GRUB_CMDLINE_LINUX_DEFAULT="quiet nosplash video=hyperv_fb:1152x864"

Save the /etc/default/grub file and update the Grub configuration:

sudo update-grub

The changes will take effect the next time you reboot the HyperV VM.


Command line options for extracting files with tar

The tar command line utility is the de facto standard in Linux for compressing/uncompressing files. Here are the most common command line options.

OptionLong OptionDescription
-x--extractExtract files from archive.
-f--fileSpecify name of file to extract.
-v--verboseList all files processed and result for each.
-j--bzip2Extract bzip2 compressed file.
-J--xzExtract xz compressed file.
-z--gunzipExtract gzip compressed file.
-Z--uncompressExtract zip compressed file.
N/A--zstdExtract zstd compressed file.

Launch tmux automatically when opening terminal/shell

If you typically run tmux immediately after opening a new terminal window or interactive shell, you can make this automatic. Just add the following lines to your .bashrc or .bash_profile.

# Start tmux automatically if interactive session and not already running
if command -v tmux > /dev/null; then
	[[ $- == *i* ]] && [[ ! $TERM =~ screen ]] && [[ -z $TMUX ]] && exec tmux new-session

Rebase local Git repository from upstream

In some instances, you will have a fork of Git repository that has significantly diverged from the upstream original. In this scenario, it is often simplest (or, at least, convenient) to rebase the local repository from the upstream. Here’s how to do this.

  1. Ensure that you have the upstream repository configured for your local repository.
git remote -v

This should display something like:

upstream        <upstream-repo-url>.git (fetch)
upstream        <upstream-repo-url>.git (push)

If the upstream repository is not displayed, then it should be added by running:

git remote add upstream <upstream-repo-url>.git
  1. Change to your main (master) branch and fetch from upstream main.
git checkout main
git fetch upstream main
  1. Overwrite the current local main from the upstream main:
git rebase upstream/main
  1. Now, you must decide if you want to merge changes from your remote main or overwrite them with the rebased update from the upstream repository. a. If you want to merge changes from your remote main:git pull origin mainYou will be prompted to merge and enter a commit message. If any merge conflicts occur, you must resolve them. b. If you want to overwrite your remote main changes:git push -u origin main –force

Your local and remote main branches forked from upstream are now synced with upstream and you can proceed with additional work against it.


The Linux Tips is a github repository by Tim Jones