Frequently Asked Questions - Unix-Linux

Connecting to cse.unl.edu and other unix servers with ssh from Linux

Open a Terminal window.  The instructions that follow are for Ubuntu Linux.  Other distributions are similar.

Click on the grid button in the lower left corner.

Enter "Terminal" in the search dialog that appears.

In the window that appears when you click on the Terminal icon shown above, enter "ssh -Y username@cse.unl.edu".  Substitue your cse user name for "username" in this command.  You can omit the "-Y" flag if you do not want to use ssh X forwarding (and if you don't know what this is, you can safely omit it).  The examples below use a user name of "sac".

The first time that you connect to cse.unl.edu, or any other Linux/Unix server, a dialog like that shown above will appear.  Answer "yes" to this.

Then enter your cse password.

Back to top ↑

Connecting to the Windows Terminal Server from Linux

The instructions that follow are for Ubuntu Linux. Other Linux distributions are similar except for installing the "Remmina" program.

Click on the grid icon at the lower left of the screen.

In the search dialog that appears enter "rdp".  The Remmina application should appear in the results. Click on that icon.

In order to connect to our terminal server you need to add a customized connection.  Click on the green "+" symbol in the Remmina program and enter the following information:

Profile name: csnt-ts (or another name that you'd like to use).

Server: csnt-ts.unl.edu

User password: your CSE password

Domain: cs.unl.edu

Color depth: True color (24 bpp)

And then click the "Save" or "Save as Default" button.

Double click on the session you just saved.

A dialog may appear asking you to accept a certificate for csnt-ts. You should accept the certificate.

Back to top ↑

Linux SMB Printing

If you encounter the error: CUPS server error "client-error-not-possible" or a similar error from the CUPS subsystem, you will need to install the smbclient package, and perhaps additionally the samba or smb-common package, using the package manager provided with your Linux distribution.

Linux Mint uses very much the same procedure except instead of being called "System Settings" the system configuration application is called "Control Center". Other Linux distributions are similar.

If you want to take advantage of more advanced printer features, such as duplexing or stapling, you can use a printer-specific driver for the printer models as seen at the bottom of Printing from Mac via SMB 10.8+ section of this FAQ

Back to top ↑

Code development on the SoC Linux servers with Visual Studio Code

If you are developing C or C++ code on the School of Computing's CSE server (cse.unl.edu) and using a Windows PC or Mac, you may benefit from using Visual Studio Code (https://code.visualstudio.com/docs/setup/windows (Links to an external site.)) as an editor. While not as full-featured as a true IDE, VS Code does give you syntax and code error highlighting which can be beneficial.

To use Visual Studio Code to edit your files stored on the CSE server, you will need to attach (mount) your CSE home directory as a remote drive. The instructions on the How do I access my home directory on campus? page will tell you how to mount your home directory as the Z: drive on a Windows system. A similar procedure for Mac allows mounting your home directory using Mac OS.

After mounting your CSE home directory, you can open Visual Studio Code to Explore your Z: drive, and open a workspace. Project directories located in your CSE home directory or subdirectories can be selected using the VS Code Explore tool. Selecting the Open Folder link in VS Code also opens the Explorer tool.

You can use Windows Explorer to copy or move source and/or project files and directories from your local laptop hard drive into your Z: drive, and create any directories or sub-directories in your CSE home directory. If you are using Git to store your project, you can clone it directly into your Z: drive by selecting the directory on the Z: drive where you want to clone the project.

To Clone a Git repository project using VS Code, choose the Clone Repository button:

Choose a directory on Z: drive to save the git project:

This example clones a project from Github, and saves the project within the directory ~/git on the CSE server. The tilde character, ~ refers to your home directory on CSE, and the git directory is an arbitrary directory that I have chosen to save all my git projects.

For MacOS, after following the instructions to mount your CSE home directory, you can use the Mac Finder application to navigate into your CSE home directory (remote location) and copy files and directories to and from your MacOS machine. Opening files and projects in VS Code is the same in MacOS using the VS Code Explorer tool, including the cloning repositories by selecting your CSE home directory remote location.

Attaching to your CSE home directory on from an off-campus network may require using a VPN connection. The VPN is a service of UNL Information Technology Services, visit the UNL ITS Global Protect VPN page for instructions on installing and connecting to the UNL network using the VPN client.

Disk Usage Considerations

The prior version of this article instructed users to load the Remote - SSH extension into VS Code and connect to the CSE server via that extension when running VS Code. The instructions above supersede those, and make the need for a .vscode-server directory in your home directory unnecessary.
If you have used VS Code using the old instructions, you may reach or exceed your quota of disk space on CSE. Using the Remote - SSH extension triggers the installation of a 100MB+ image of the VS Code Server into your home directory. With prolonged use of VS Code and Remote - SSH, you may have multiple copies of the VS Code Server application being placed in a subdirectory of your home directory. This is caused by VS Code and the update mechanism, resulting in extra copies of the executable files in ~/.vscode-server/bin. These old/extra copies count against your disk quota on the CSE server, and may result in you receiving disk quota exceeded emails from CSE Systems Support. These old/extra copies can safely be removed, an easy way to tell is if the date reported by ls -lt is the most recent, that is the copy to keep. All older copies can be safely removed.

Below is an example of what you will find in the VS Code Server directory, ~/.vscode-server and how to remove extra copies of VS Code executable files:

wmotycka@cse:~> ls -l ~/.vscode-server/bin/
total 12
drwxr-xr-x 6 wmotycka staff 4096 Oct 26 07:54 d2e414d9e4239a252d1ab117bd7067f125afd80a  # NOTE OLDEST DATE
drwxr-xr-x 6 wmotycka staff 4096 Jan 16 08:55 ea3859d4ba2f3e577a159bc91e3074c5d85c0523
drwxr-xr-x 6 wmotycka staff 4096 Nov  8 09:00 fcac248b077b55bae4ba5bab613fd6e9156c2f0c    # NOTE OLD DATE
wmotycka@cse:~> cd ~/.vscode-server/bin
wmotycka@cse:~/.vscode-server/bin> du -hs
317M    .
wmotycka@cse:~/.vscode-server/bin> du -hd1
112M    ./d2e414d9e4239a252d1ab117bd7067f125afd80a
104M    ./ea3859d4ba2f3e577a159bc91e3074c5d85c0523
103M    ./fcac248b077b55bae4ba5bab613fd6e9156c2f0c
317M    .
wmotycka@cse:~/.vscode-server/bin> cd ../..
wmotycka@cse:~> # Remove the extra/oldest copy/copies of the VS Code Server using 'rm -rf'
wmotycka@cse:~> rm -rf ~/.vscode-server/bin/d2e414d9e4239a252d1ab117bd7067f125afd80a
wmotycka@cse:~> rm -rf ~/.vscode-server/bin/fcac248b077b55bae4ba5bab613fd6e9156c2f0c
wmotycka@cse:~>

The last two commands above remove the oldest copies of VS Code Server images while retaining the most recently dated copy present.

Latest Notes

Unless you have a particular need, we recommend completely removing the ~/.vscode-server directory and instead access files on CSE using the method above. The ~/.vscode-server directory is necessary for operation using the Remote - SSH extension but is unnecessary when using VS Code on remote-mounting files to your personal Windows or Mac machine. There is also a performance improvement to VS Code by using remote-mounted files instead of the Remote - SSH method since all of the editing work is done on your local system. Disk usage reduction and performance improvement together justify changing to the method described above. If you want, remove all the contents of the ~/.vscode-server/bin/* and then use the instructions above to connect to your project. The only change you should see may be the need to install extensions locally that you originally had installed on the CSE server.

The Remote - SSH extension can also cause performance issues on the CSE server. Features of the extension like file search spawn processes that can linger in the CSE process table, becoming "zombie" processes that can only be removed by manual intervention (reboot or using 'kill' on each.) These lingering processes reduce available processing time, which you experience as slow compiles and program execution.

VS Code on a Linux client

Running VS Code on a Linux client (for example running Ubuntu on your own laptop) the way you mount the drive is important. You will need to use the fuse filesystem and the base SAMBA client packages samba-client and samba-common. For more specifics follow (or search) the instructions for your Linux distribution regarding installing SAMBA client. For my installation (Kubuntu 22.04), I also installed samba-common-bin, samba-libs, samba-vfs-libs, smbclient, cifs-utils, and winbind packages.
The Gnome Virtual Filesystem gvfs facility, if properly configured, will mount your CSE home directory to one of the two the default locations, ~/.gvfs or /run/user/<user-number>/gvfs. This gvfs facility is easy to use from the Gnome Files application, but doesn't work properly with all file managers like Dolphin, where the mounted CSE filesystem is not being exposed into the ~/.gvfs or /run/user/<user-number>/gvfs directory, as the gvfsd-fuse documentation describes. Using the Gnome Files application, the /run/user/<user-number>/gvfs directory contains the mount point directory that allows access to your CSE home directory. Later instructions provide the Linux shell command-line necessary to mount your home directory without the use of a file manager application.
After installing the appropriate packages, the following steps will mount your CSE home directory:

• Launch the Files application, and select Other Locations from the left-side menu:

  

  In the Connect to Server field enter the URL for connecting to the the Samba server, smb://cse-smb1.unl.edu/%lt;cse-login%gt; where cse-login is your username on the CSE server. Finish this step by pressing the Connect button.
• You will next be prompted to enter your CSE username (login id), network domain, and password. Enter cs.unl.edu for the Domain, and your CSE username and password:

  

• Your CSE home directory is now mounted and can be accessed from within VS Code using the Explore Open Folder action:

  

You can also mount your CSE home directory using the Linux command-line shell. The following outlines the basic steps:

• Open a terminal window and execute the command: gio mount smb://cse-smb1.unl.edu/cse-login replacing cse-login with your login id on the CSE server
• You will be prompted for you CSE login id and credentials:

  

  If your login credentials on your Linux client (laptop) are not the same as your CSE credentials, enter your CSE login username to the User prompt. The Domain is cs.unl.edu as shown in figure 4. Enter your CSE password to the Password prompt. If successful, the shell command prompt is returned.
• You can then create symbolic links from the mount-point directory under /run/user/<user-number>/gvfs/ into a place in your home directory that will help navigate to it:
  ln -s /run/user/1000/gvfs/smb-share\:server\=cse-smb1.unl.edu\,share\=wmotycka ~/Documents/CSE
  creates a symbolic-link (directory path), /home/wmotycka/Documents/CSE, to the CSE home directory of the user wmotycka that can be accessed from within the VS Code Explorer, Open Folder action.

  

• Projects and source files that require compilers like C and C++ source will still need to be compiled from a shell command-line on CSE, since the compiler and libraries on your local Linux client system may not be compatible with the libraries installed on CSE. This does not preclude compiling and running your code on your local Linux client system, which you can do for testing and development.
• Project management tools like git have extensions in VS Code that allow managing shared code projects and cloud-based source repositories through the VS Code application. The CSE server supports git and maven (mvn) source management as well as svn source version control. Code-base management can be performed from both within the VS Code application and using CSE server shell commands.

Back to top ↑

Compilers and Interpreters available on cse.unl.edu

Compilers

gcc	GNU CcCompiler
g++	GNU C++ compiler
javac	Java compiler
gfortran	GNU Fortran compiler
clisp	Common Lisp
alisp	Allegro Common Lisp

Interpreters

perl	Perl command line scripting language
php	Command line and web scripting language
ruby	Command line and web scripting language
pyton	Command line scripting language

Back to top ↑

Basic Unix Commands

Note on systax used in document:

items in [] are optional

items in {} can be repeated 0 or more times.

fn = filename (possible with wildcards)

dir = directory name

File System:

cat fn {fn}

concatenate, can be used to display a file.

more fn {fn}

displays file(s) one screenful, at a time. At MORE prompt enter: space bar - for next screenful <cr> - for next line, q - to quit more command.

cp fn1 fn2

copy file - fn1 to fn2, fn1 remains.

mv fn1 fn2

move (rename) file - fn1 to fn2, fn1 no longer exists.

rm fn

remove (delete) file.

lpr -Pprinter_name fn

print a file, printer_name = ps16 or ps17.

ls [fn]

list files in current directory option filename is usually used to display files matching wildcards. (ie. ls *.s will list all assembly language source files.)

ls -lgRa

Recursive listing of all files within all subdirectories. Longest, most descriptive format.

cd [dir]

change working directory, ``home directory" if not specified. cd .. will backup one level in the directory structure.

pwd

print working directory path.

mkdir dir

make a new directory, giving it the name dir (within the current directory)

rmdir dir

remove (delete) the directory named dir (the directory must be empty)

file fn

Tries to determine what kind of information is in a file by consulting the file system index and by reading the file itself.

du

Prints a summary of total space occupied by all files in a hierarchy.

sort fn

Sorts ASCII files line-by-line.

chmod

Change file permissions. See the man pages for a complete description of settings. Use this command to determine whether files can be readable by other people.

find . -name filename

In it's most simple form find will display all files in this and any subdirectories which name matches filename. See the man page for find for a complete description of this command. Find can be use to find files matching name patters, dates, owners, sizes, etc. and to perform action on matched files.

gzip fn

Use this to compress files not needed immediately to conserve disk space. A ``.gz" will be appended to the filename.

gzip -d

Uncompresses files with ``.Z" at end of filename.

head {-n} fn

Prints first n lines of a text file. If no number is specified, the default is 10.

tail {-n} fn

Prints last n lines of a text file. If no number is specified, the default is 10.

tail {-n} fn

Prints last n lines of a text file. If no number is specified, the default is 10.

grep string fn

Print occurrences of string in file

Remote Connection:

ssh remote_host

Launch a secure shell to the remote host specified.

Communication:

finger

list users currently on the system.

who

list users currently on the system.

w

what, similar to who but includes what they are doing.

pine

Email message utility.

pine username

will send Email to username.

talk username

allows two users to have an interactive conversation over the computer.

write username

allows a user to put a message on another users screen.

mesg n

Makes it so other users cannot interrupt you with talk requests, and write messages.

Help:

man command

manual, returns the manual for the given command from the UNIX on-line manual. If you cannot find a man page for a command, there may be one in a directory that isn't in your MANPATH. Your MANPATH is setup in the .login file and a command's man page is usually in a man directory off the same parent directory as the command. For example, if the command is located in /usr/local/bin, its man page is probably in /usr/local/man.

man -k keyword

display manual headings containing the word keyword.

Miscellaneous:

alias name def

Assigns ``def'' to the alias ``name''. Usually found in the .cshrc file. For example ``alias dir ls -la''.

clear

Clears screen.

history

List history of commands used.

!!

re-execute the last command entered.

!x

re-execute the last command that started with the given letter, eg !c will re-execute the last command that began with the letter c.

jobs -l

list all currently stopped or background jobs with job numbers and process ID's.

kill [-signal] %jobnumber

kill the given job number. Use the "jobs" command to show running jobs. You might need to kill a job to logout if you have accidentally put a job into the background or stopped a job. Use -TERM or -KILL singlas to force quite jobs.

kill PID

kill the given Process. refer the "ps" command to find running processes. Use -TERM or -KILL singlas to force quite jobs.

ps -fa

list all your current processes. See the man page for a full description.

<control>C

kill the current process - the one in the  foreground

<control>Z

suspend foreground process.

bg

put a suspended job in the background.

fg

put a suspended or background job in the foreground.

date

display time and date.

To find out about more commands, click  here.

Back to top ↑

MySQL Database Account — Requesting and Connecting

MySQL is the general purpose database subsystem provided by the School of Computing (SoC) for student use. Each user is allowed one database. This database is automatically created after requested by a user and is named the same as your SoC systems username. Users can create as many tables as they want but we ask that they keep all databases to a reasonable size.

Requesting a MySQL Account and Database
Any user with a SoC account can request access to the MySQL server running on the cse-linux-01.unl.edu server.  Users with legacy CSE server accounts can request access to the MySQL server running on the cse.unl.edu server.  To determine if you already have a MySQL account, log in to the SoC Account Management Utility (AMU) page at https://cse.unl.edu/account with your CSE account credentials. Once you log in, click the Account Settings link on the left page. If you have already been granted a MySQL account on one of the SoC servers (cse.unl.edu, cse-linux-01.unl.edu, or nuros.unl.edu) you will see a message indicating on which system(s):

If you do not have any MySQL accounts showing on Account Settings, you will need to send an email to support@cse.unl.edu requesting a MySQL account.  If you already have an account on the cse.unl.edu server, you may request a MySQL account on that system, otherwise your MySQL account will be created on the cse-linux-01.unl.edu server.

The MySQL credentials you receive will have a login that matches your CSE server (cse.unl.edu) username, but your password for the MySQL server is completely different from your CSE server password.

Accessing the MySQL server is done by one either logging into the cse.unl.edu server using your CSE username and password, or connecting to the cse.unl.edu server using a TCP/IP connection.

Connecting to MySQL Database
To connect to a MySQL database from the command-line on cse.unl.edu, type mysql -p and hit enter. The password you will be prompted for will be your MySQL password, which is different from your cse.unl.edu server password. Enter the MySQL password you received in the email message that was sent when you requested your account. You should be presented a MariaDB [(username)]> prompt from the MySQL application.

To connect to a MySQL database from the command line from another computer with the MySQL command line client installed. Run 'mysql -h cse.unl.edu -u YOUR_CSE_LOGIN -p' and hit enter. Note: the password you will be prompted for will be your MySQL password, which is different from your SoC password. MySQL passwords can be reset using the SoC Account Management Utility page.

TCP/IP connectivity requires authentication using your MySQL credentials and is supported on the standard MySQL service port, 3306, on the CSE server.

Back to top ↑