- Prerequisities
- Get the sources
- testing the distribution
- Installing the software
- Post installation steps
- Links
Install python >= 2.2.1, and a PostgreSQL or Oracle database. Install pyPgSQL >= 2.2 or DCOracle2.
- Debian packages: python2.2 python2.2-dev python2.2-xmlbase postgresql python2.2-pgsql
- RedHat packages: ???
Create a database instance where cerebrum should be ran, as well as a user which cerebrum should connect as.
Instructions for creating the Oracle users:
CREATE USER cerebrum IDENTIFIED BY somepassword;
GRANT connect, resource, create role TO cerebrum;
CREATE USER cerebrum_user identified BY somepassword;
GRANT connect, resource TO cerebrum_user;
Instructions for creating the PostgreSQL database and user:
createuser cerebrum
createdb -E unicode cerebrum
Make sure that the database user 'cerebrum' has proper access to the new database.
git clone https://utv.uio.no/stash/scm/crb/cerebrum.git
Set PYTHONPATH to point to the directory where you extracted the sources. Edit cereconf.py and set DB_AUTH_DIR, CEREBRUM_DATABASE_CONNECT_DATA['user'], CEREBRUM_DATABASE_NAME and CLASS_DBDRIVER.
Warning: the target datbases cerebrum tables will be dropped and recreated when running this script!
Run make fullcheck
.
If any of the tests fail on your system, you chould investigate the error before proceeding with the installation.
Enter the directory where you installed the sources, and run:
./configure --prefix=/cerebrum
make
make install
This will install the Cerebrum python modules into your site-packages directory. A number of other files are placed as documented in the top of setup.py. To install to an alternative location, use the --prefix option to install. If you use --prefix, remember to set PYTHONPATH when running scripts.
Run make bootstrap
to create the required tables.
- cronjobs
- auto starting bofhd
- ?