Changing ttlpv.sql to use insert soft (and also making sure to use the
stored id) appears to fix the problem I am encountering.
I prepared a modified ttlpv.sql (attached) and reinstalled Virtuoso as
follows:
cd /home/virtuoso/test
killall virtuoso-t
rm -rf vos
cp ttlpv.sql ./virtuoso-opensource/l
Aah. It's likely then that something in MacOS doesn't switch between threads
within ttlpv.sql.
It is also possible that there is a similar issue in ttlpv.sql having to do
with type ids (rdf_rl_type_id). That code is very similar to the code where I
found the bug.
peter
On 12/19/18 11:05 AM, H
Hi Peter,
My original testing was on Darwin (macOS), I have switched to CentOS 7 and have
been able to recreate the issue on first run:
[hwilliams@localhost database]$ sh test.sh
Wed 19 Dec 15:43:33 GMT 2018: 3/ Start up Virtuoso with an empty database
OpenLink Virtuoso Interactive SQL (Virtuos
That's not my experience on several machines. I've experienced this bug on
two servers, one running CentOS release 6.9 (Final) and one running Ubuntu
16.04.5 LTS. The bug depends at least on loading in parallel *and*
encountering new language tags. Loading the files a second time will almost
cer
Hi Peter,
I generated the datasets from your python script and loaded them into a local
Virtuoso open source multiple times but did not see any occurrences of the
error:
SQL> select * from load_list;
ll_file
ll_graph
I created some synthetic data that tickles the bug reliably on my machine with
a standard virtuoso.ini (just adding the directory for the files to the
allowed list). I'm attaching the generator program for the files and a
loading script.
peter
On 12/18/18 9:46 AM, Peter F. Patel-Schneider wrote
I did a bit of digging and it sure looks as if there is a race condition in
rdf_rl_lang_id in ttlpv.sql. This code appears to check to see if the
language tag is already in DB.DBA.RDF_LANGUAGE and adds it if not. But
another thread could do the same insert between the check and the insert, as
fa
I'm loading the Turtle Wikidata RDF complete dump, split into pieces and
loaded with 10 active readers. About half the time the load fails with one
or more of these errors. The errors are always near the beginning of the
load---in the first group of 10 files to be loaded and near the beginning o
Hi Peter,
The triple value do indeed appear to be valid, but the problem could be
somewhere else in the dataset file and not necessarily on the reported line or
line before it.
Is it a public dataset you are loading and if so can you provide a copy for
local testing ?
Best Regards
Hugh Willia
