Who is Cartesius?

The system is named Cartesius in honour of René Descartes (Cartesius in Latin), a French philosopher, mathematician, and writer who spent most of his adult life in the Dutch Republic.

A comprehensive description of Descartes and his work can be found here.

My job doesn't start with a status 'ReqNodeNotAvail'

This usually happens when a maintenance session is planned. You can see planned maintenance on the system status page or in the message of the day when logging in on Cartesius.

Jobs with a maximum wall clock time longer than the time until the start of the maintenance, will not start until after the maintenance and are indicated with a status 'ReqNodeNotAvail' in the squeue output.

A workaround is to use a shorter maximum wall clock time.

Help, I can't login! Is my account blocked?

There can be different reasons why this happens:

• One of the most common reasons is that the system is in maintenance. You can see planned and ongoing maintenance on the system status page.
• Cartesius uses a white-list, so you might find that the system cannot be accessed while travelling. For these moments, please use the doornode.
• If you need access from a location that you will use regularly and long-term, please contact the helpdesk andd fill in your external IP address. You can easily find your external IP address by visiting http://www.whatismyip.com.

I need to see the output of my batch job immediately while executing or my program crashed but the output seems cut off!

If your program is compiled with the GNU gfortran compiler, set the following environment variable:

export GFORTRAN_UNBUFFERED_ALL=y

For C programs, the buffering can be changed using the command setvbuf. E.g. standard output can be unbuffering using:

#include <stdio.h>
...
setvbuf(stdout, NULL, _IONBF, 0);

I can't use CVS.

Cartesius does not support the default remote shell 'rsh' for security reasons. Please use:

export CVS_RSH=ssh

How can I determine the memory usage of my application?

The SLURM batch scheduler logs the memory usage of your application and it can be retrieved after your job has ended:

$ job-statistics -j 1155623
...
              AveRSS :  11077K
              MaxRSS :  11576K
          MaxRSSTask :  46
          MaxRSSNode :  tcn828
...

it shows the average and maximum memory use per MPI task, which MPI task used the maximum memory and on what node.

If you want to print the memory usage as part of your application, note that the linux system call getrusage() isn't fully implemented under linux, see 'man 2 getrusage'. Please compile and use the C routine printmem(), which prints the memory usage.

Which nodes are allocated to my job?

The environment variable $SLURM_NODELIST contains the names of the nodes. The format is something like: tcn[9006-9008]. Using the program scontrol you can obtain the nodenames, one name per line:

$ scontrol show hostname $SLURM_NODELIST
tcn9006
tcn9007
tcn9008

Using the command nodeset, you get all nodenames on one line:

$ nodeset -e $SLURM_NODELIST
tcn9006 tcn9007 tcn9008