Table Of Contents

Previous topic

CORK and BPR Field Guide

Next topic

Manpages

This Page

CORK Computer Setup

Required software

A computer running Linux is the simplest platform to install and run the CORK software on. It is also known to run on Mac computers and most data data processing can be done on Windows machines, but mlterm the programm used to communicate with CORKs and download the data does not work under windows and the necessary development environment to compile various C-programs might be challenging to setup on Mac.

Check out the CORK software repository

CORK software, calibration and parameter files are maintainded at the CORK Observatory Software Repository.

It is best to have svn installed to check out the code. Check if svn is installed by running svn --help. If there is not help displayed you have to install svn e.g. by running sodu yum install subversion.

Now, go into the home directory of the cork user (cd ~) and run

svn co https://svn.code.sf.net/p/corkobservatory/code/ corkobservatory

Note

SourceForge changed their naming scheme. Do not use the following anymore: svn co https://corkobservatory.svn.sourceforge.net/svnroot/corkobservatory corkobservatory

If you are stuck at revision 65 after update Updated to revision 65.

svn info URL: https://corkobservatory.svn.sourceforge.net/svnroot/corkobservatory

svn switch --relocate https://corkobservatory.svn.sourceforge.net/svnroot/corkobservatory https://svn.code.sf.net/p/corkobservatory/code/

After update revision should be greater than Revision: 73

This will create a corkobservatory directory that contains program source codes, calibration information, etc. To update these files to the latest version you can cd into the corkobservatory directory and run svn update at any time when you have an internet connection.

Compile and install mlterm, mlbin, mldat9

mlterm, mlbin and mldat9 are the core utilities (programmed in C) to setup CORKs and download the data, strip unneccessary filesystem information from the downloaded *.raw files, and to calibrate the data contained in the *.bin files, respectively. In some special cases, when there are problems with data consistency or with special instrument setups you need to use some Python scripts, that will be discussed later, instead of mldat9.

Since the core utilities are supplied as C source code, you need to have a C-compiler and basic development tools installed. cd into the mlterm, mlbin, and mldat directories under ~/corkobservatory/C-tools and run

  1. make
  2. sudo make install

in each directory to compile and install the programs individually.

NOTE: there is also a variant of mlterm mlterm_vpn that gets automatically installed with mlterm that is for use with instruments connected over connections with high latency (e.g. TCP/IP, VPN,...).

Install RS-422 adapter driver if necessary

Many USB/RS-422 adapters run under Linux out of the box, but the Moxa UPort-1130, which we found is the most reliable adapter, requires the compilation/installation of a seperate driver. We had problems in the past, because the driver was not compatible with current Linux distributions, but fortunately, Moxa did update the driver in summer 2012.

So if you want to use a Moxa adapter make sure to download and install the latest diver either from their website or use the copy that is stored in the software repository (~/corkobservatory/ComputerSetup/drivers/MoxaUPort). Follow the instructions in the readme.txt file under “Module driver configuration”.

If the driver install succeeded running mlterm with the Moxa UPort plugged into a USB port should start an mlterm session with mlterm trying to connect to the instrument. If the adapter is not recognized by the OS mlterm will fail with the following message:

/dev/ttyUSB0: No such file or directory

Install dependencies for Python scripts

The Python scripts need

  1. Python (tested with 2.6 and 2.7)
  2. numpy
  3. matplotlib (optionally for plotting)

So first check if Python is already installed python --version and install it with e.g. sudo yum install python if not. Now you can try to install the numpy and matplotlib packages of your distribution e.g.:

  • sudo yum install python-numpy
  • sudo yum install python-matplotlib

or install easy_install e.g. sudo yum install python-setuptools and then run:

  • sudo easy_install numpy
  • sudo easy_install matplotlib

Now, cd into the ~/corkobservatory/ComputerSetup/scripts directory (see Check out the CORK software repository for how to set it up) and run ./createPythonToolsSymLinks.bash to add symbolic links to the python scripts dumpBin.py and calibrateLogfile.py to your path. If everything worked out, running dumpBin.py --help and calibrateLogfile.py --help will display the respective help information.