HI Vanessa,
I installed the latest version of virtuoso on my Mac Lion. I
modified virtuoso.ini to use up to 4GB of memory (I have 8GB),
however when I am uploading the DBpedia infoboxes it fails after a
couple of hours or more.
00:48:36 PL LOG: Loader started
00:48:36 PL LOG: No more files to load. Loader has finished,
00:48:59 PL LOG: Loader started
00:48:59 PL LOG: No more files to load. Loader has finished,
00:51:01 PL LOG: Loader started
01:42:24 Checkpoint started
01:42:34 Checkpoint finished, log reused
02:42:37 Checkpoint started
02:42:50 Checkpoint finished, log reused
02:49:45 Server received signal 11. Continuing with the default
action for that signal.
I use the DBpedia script and I did not have any problem to install
and upload virtuoso in Linux using the same procedure. Any idea what
the problem can be?
cat BulkLoaderScript.txt | /usr/local/src/virtuoso/bin/isql
SQL> /usr/local/src/virtuoso/bin/isql
SQL> ld_dir ('/virtuosodata/dbpedia/', 'infobox_properties_en.nt', 'http://dbpedia.org
');
SQL> rdf_loader_run();
To be able to analyze what is going on, you need to setup your system
to produce meaningful core dumps on Mac OS X.
Firstly i suggest you run the following commands to re-configure
virtuoso build to include debug info:
./config.nice --with-debug
make clean
make
Next you run the following command so your system will write core dumps
ulimit -c unlimited
and then start the virtuoso server from this same shell and run your
bulkload script.
When it next gets a signal 11, you will find a core in /cores/core.PID
where PID is the process id of your virtuoso server.
Then you can do:
gdb virtuoso-t /cores/core.PID
and on the gdb prompt run the
bt
command to show the stack trace for this crash.
Please send me the output of this command, plus the last couple of
lines of your virtuoso.log so i can analyze what is going on and what
steps to take next.
Patrick
