How to Set Up a Unix Shell
This is an introduction to setting up your account on our systems. When first logging in, you are presented with a default set-up that enables the use of basic system commands, simple compilers, and access to the scheduler. This help file is meant to explain how to modify that default.
A shell is a command line interface that lets users issue commands to the operating system. It contains the basic Unix commands, as well as many more adcanced ones. The default shell for a Linux system is bash ("Bourne again" shell). There are others:
- csh (C shell)
- ksh (Korn shell)
They differ in their "feel" and how they handle system commands. ksh and bash largely understand each other's command syntax. csh is somewhat different.
bash is a "GNU" shell that uses features from the older shells csh and ksh. It is quite popular because it offers many convenient features, such as command and file-name completion by pressing the "Tab"-key. Its scripting syntax is almost identical to the one of ksh, which enables you to use ksh set-up files.
Some people prefer csh best because it is simple and its syntax is similar to the C programming language (thus the name). We discourage its use (see this for reasons).
ksh offers some extended features but is rarely used these days as it is similar to bash but offers less features.Some programs require a specific shell to run. You can always call a shell by typing
/usr/bin/name_of_shellwhere name_of_shell might be csh, tcsh, ksh, bash or sh (the latter is another standard shell that is often used for scripts). You may have to maintain set-up files for the shells you use, even if they are not your login shell.
In scripts, shells are usually called with a "hash-bang" sequence, that looks like:
#!/bin/bashin the first line. In such cases, it can happen that environment variables, such as PATH get reset to some default. This is why it may be necessary to use the -f option:
#!/bin/csh -fwhich avoids sourcing in the set-up files, leading to consistent behaviour for different users. All shells are documented in the man pages: type
man name_of_shellto get more information than you'd care to know. Here are links for pretty complete documentation for the shells online:
In the following section, we assume a bash shell.
Setting up a shellWe are using a system called usepackage to setup the shell. This allows replacing the setting of several environment variables, and the execution of set-up scripts by simple commands of the form
use packagewhere package stands for a pre-defined application or feature that is to be included. We discuss the use of use in the next section.
There are several set-up files in your home directory:
- .bashrc is "sourced in" every time a bash shell is invoked.
- .bash_profile applies only to login shells, i.e. when you access the system from outside.
usepackage and the use commandMost set-ups involve the setting of environment variables. These are variables that are used by your shell to determine its desired behaviour. Other set-up operations may include the execution of scripts. To simplify this task, our system supplies the use command. Instead of setting each variable and execute each script separately, everything is "lumped in" to the command
use package_namewhere "package_name" stands for an application or functionality that you want to include. A list of possible package_names can be obtained by typing
Here is an example: To be able to use the electronic-structure software Gaussian, you need to execute ("source in") a lengthy set-up script, and set a few environment variable, e.g. to tell the system where to place temporary files:
export g09root = "/opt/gaussian/g09e1" . /opt/gaussian/g09e1/g09/bsd/g09.profile export GAUSS_SCRDIR = "/scratch/$LOGNAME"These commands can be replaced by
use g09and the above commands are stored in a file
/opt/usepackage/etc/usepackage.confThe use command also works inside of your login set-up file, i.e. .bash_profile, so you won't have to retype it every time you log in. By default, users are supplied with a basic set-up that corresponds to
use standard-user-settingsand includes access to most system functions, scheduler, and some compilers. Note: New shells do not always know about the use command. In that case you either have to set up that shell "manually", or teach it about use. This can be done with a single line:
Note that it is rarely necessary to make invoked shells aware of use, as they inherit the setup of the invoking (login) shell.
You can of course apply settings directly without using "use".
One of the most important environment variables is PATH, which tells the system where to look for the commands you issue. You may want to make your shell aware of some directories with system commands and shell commands in them.
Another environment variable that is often useful is MANPATH. This is for the Unix manual pages, and tells the system where to look for online-documentation.
Yet another one is LD_LIBRARY_PATH, which is sometimes used by applications to find dynamic runtime libraries. If you experience problems with missing libraries, try playing with LD_LIBRARY_PATH, otherwise it's best left unset.
The command to set an EV is the binary operator '='. This is often followed by the export command, which makes the variable part of the environment:
VARIABLE=VALUE export VARIABLENote that it is possible to place "export" in front of the variable assignment instead of issuing two separate commands:
export VARIABLE=VALUETo access the value of a environment variable, place a "$" in front of it. For example you want to see which value your variable PATH has, type
echo $PATHwhere "echo" is a standard Unix command, and "$PATH" returns the value of PATH. The following command will append something to a previously defined variable:
Here, "PATH" is the variable and "$PATH" is its present value.
Sometimes a variable needs to be reset for a specific application. It is then best to write a shell script that sets the variables and starts the application, rather than setting the variables globally in your start-up files. In this example:
The variable OMP_NUM_THREADS (which determines how many threads are being used) is set only for this specific run of the multi-threaded program omp_program.
You can consult the configuration files of "usepackage" /opt/usepackage/etc/usepackage.conf to find out which setting you need to apply to run a specific software or access certain features. The syntax in that file is not hard to read, for instance the entry
>> g09 : "Gaussian 09 Update E.01" << g09 : g09root = "/opt/gaussian/g09e1", <[ . /opt/gaussian/g09e1/g09/bsd/g09.profile ]>, GAUSS_SCRDIR = "/scratch/$LOGNAME";
is responsible for the settings in the "Gaussian" example in the previous section.
Running pre-installed software
A lot of software is pre-installed on our clusters. Some of this software requires specific license agreements, other programs are freely accessible. With the use command, most of them can be set up with a single line such as "use fluent" in your shell's start-up file. If the software you want to run is not included in our usepackage list, please contact us, and we can include it. If you are using very specific software that is not accessed by other users, you might have to do the setup manually.
Here is a few steps to follow in that case.
- Check out the documentation for the specific program, including users' manuals and home pages.
- Inform yourself about licensing. Some software requires each individual user to hold a license, some is covered by a collective license agreement, some does not require a license at all. For example, the finite-element structural code "Abaqus" is only accessible to users who work at an institution that is covered by a local license, whereas the license agreement for the electronic-structure code "Gaussian" covers all our HPCVL users. Finally, code such as "Gamess" (another quantum-chemistry program) are free to use by all users, although the distributor encourages registration.
- Set the proper environment variables. This can usually be done in your shell setup files, since you'll be running the same code on most occasions you log on. These variables might include the PATH, but also variables specific for the program in question. Which ones to set you will be able to find out in most cases from the program documentation. Remember that this is only necessary if no entry exists in the "usepackage" configuration file, which can be checked by running "use -l".
How do I run parallel code ?
That depends on how the code is "parallelized":
- If it was "multi-threaded" by the compiler (automatic or via compiler directives), it is usually enough to set the environment variable PARALLEL or OMP_NUM_THREADS to the number of threads that should be used.
- If it is MPI code, a special parallel runtime environment has to be used. The command there is mpirun, which has command-line options that let you tell how many and which processors to use. This command is part of the Cluster Tools parallel runtime environment. Cluster Tools involves a good deal of commands that let you modify the condition under which your program runs. The settings for these are included in our default setup.
You can learn more about parallel code by having a look at our Parallel Programming FAQ. We also have a bit more specific information about parallel programming tools, namely OpenMP compiler directives and the Message Passing Interface (MPI).
How do I get my terminal to work with this system ?
In most cases your terminal settings are automatically selected and work "as is".
In some cases when the machine used to connect to our machines is non-standard, an environment variable has to be set to tell our system what kind of terminal you are using. This variable is TERM. For instance, if you log in from a non-standard terminal type such as "vt220" you may have to type (bash):
- export TERM=vt100
If you are always logging in from the same terminal, you might as well put this into your setup file. In most cases you don't have to issue this at all because the system figures it out by itself.
If you want to run graphics applications that create windows on your desktop you will on rare occasions have to set another environment variable called "DISPLAY". That is done like:
- DISPLAY=ip_address:disp_num; export DISPLAY
for the others. ip_address stands for the IP address of your terminal or desktop, and disp_num is its "display number" (which is usually 0). If you don't know it, you can usually find it by typing "who am i" on sfnode0. Setting this environment variable will then display any newly called interfaces on your terminal. Or will it? ... Not quite. First you have to issue a command "xhost" on your local machine (usually a desktop). Typing "xhost sflogin0.hpcvl.queensu.ca" on your terminal enables you to receive displays from this remote machine. Now you're ready.
Note that the above procedure of setting the DISPLAY environment variable can most often be avoided by configuring your connection software (see next section about ssh clients) to enable "X11 forwarding".