tim's journal

Installing boinc

Here we go, here we go!

1. Download the latest boinc tarball
   
2. Though technically not needed, I upgraded automake from 1.6.x to 1.7.8 via rpm

3. Now, gingerly follow the directions from Berkeley... I ran

   ./configure

   got a whole bunch of positive looking messages, and then

   Traceback (most recent call last): File "test/test_sanity.py", line 5, in ? from testbase import * File "test/testbase.py", line 16, in ? from Boinc.setup_project import * File "/root/src/boinc-2.04/py/Boinc/setup_project.py", line 9, in ? from Boinc import database, db_mid, configxml, tools File "/root/src/boinc-2.04/py/Boinc/database.py", line 38, in ? import MySQLdb, MySQLdb.cursors ImportError: No module named MySQLdb

   Yikes, that's no good. RPM check shows that I have installed MySQL-python.
4. Turned out the rpm was bad, so uninstall the MySQL-python rpm, download the source code, and

   $ tar xfz MySQL-python-0.9.2.tar.gz $ cd MySQL-python-0.9.2 $ python setup.py build $ su # python setup.py install

5. No, still problems

   Traceback (most recent call last): File "test/test_sanity.py", line 5, in ? from testbase import * File "test/testbase.py", line 18, in ? import cgiserver ImportError: No module named cgiserver

6. It turned out that the latest version I had (2.04) was missing cgiserver.py - I ended up, with some advise from Karl Chen, downloading the latest nightly tarball. That fixed the the problem, and allowed me to pass test/test_sanity.py.
7. Type make, and, if your setup is like mine, it'll complain about not finding gl.h. You need to add the developer XFree86 rpm, which requires fontconfig-devel, which requires freetype-devel - these can be found (at least for RH9) here- don't bother getting the latest version, there are too many dependencies. Here's what I used

   freetype-devel-2.1.3-6.i386.rpm fontconfig-devel-2.1-9.i386.rpm XFree86-devel-4.3.0-2.i386.rpm

   It should compile at this point.
8. Haha, it'll compile, but 1/2 tests will fail. (Try 'make check'.)

   -- Testing standard upper_case application Posted by torque at October 13, 2003 1:46 PM | TrackBack