Hi Jerven, In Virtuoso 7+ the RDF_QUAD table is created with COLUMN (column wise) indexes by default, whereas you seem to have been creating additional BITMAP (row wise) indexes , which is not recommended ie mixing column and row wise indexes as this can result in the unpredictable behaviour you are experiencing, see:
https://virtuoso.openlinksw.com/dataspace/doc/dav/wiki/Main/VirtRDFPerformanceTuning#RDF
Index Scheme
<https://virtuoso.openlinksw.com/dataspace/doc/dav/wiki/Main/VirtRDFPerformanceTuning#RDF%20Index%20Scheme>
Is there a reason why these additional indexes were being tried as the 5
default (2 full & 3 partial) column wise indexes should suffice for most use
cases, with only one exception found to date which is detailed in the Index
Scheme Selection section of the Wiki doc:
https://virtuoso.openlinksw.com/dataspace/doc/dav/wiki/Main/VirtRDFPerformanceTuning#Index
Scheme Selection
<https://virtuoso.openlinksw.com/dataspace/doc/dav/wiki/Main/VirtRDFPerformanceTuning#Index%20Scheme%20Selection>
Best Regards
Hugh Williams
Professional Services
OpenLink Software, Inc. // http://www.openlinksw.com/
<http://www.openlinksw.com/>
Weblog -- http://www.openlinksw.com/blogs/ <http://www.openlinksw.com/blogs/>
LinkedIn -- http://www.linkedin.com/company/openlink-software/
<http://www.linkedin.com/company/openlink-software/>
Twitter -- http://twitter.com/OpenLink <http://twitter.com/OpenLink>
Google+ -- http://plus.google.com/100570109519069333827/
<http://plus.google.com/100570109519069333827/>
Facebook -- http://www.facebook.com/OpenLinkSoftware
<http://www.facebook.com/OpenLinkSoftware>
Universal Data Access, Integration, and Management Technology Providers
> On 13 Feb 2017, at 08:49, Jerven Tjalling Bolleman <jerven.bolleman@sib.swiss
> <mailto:jerven.bolleman@sib.swiss>> wrote:
>
> Hi Hugh,
>
> I managed to reproduce.
>
> PREFIX up:<http://purl.uniprot.org/core/ <http://purl.uniprot.org/core/>> ASK
> {GRAPH {?subject a
> up:Protein . ?subject up:annotation ?target .}}]
>
> -> SQ156: Internal Optimized compiler error : sqlo table has no index in
> sqldf.c:3777. Please report the statement compiled.
>
>
> SQL> SET SPARQL_TRANSLATE ON;
> SQL> sparql PREFIX up:<http://purl.uniprot.org/core/
> <http://purl.uniprot.org/core/>> ASK {GRAPH
> <http://sparq.uniprot.org/uniprot <http://sparq.uniprot.org/uniprot>>
> {?subject a up:Protein . ?subject
> up:annotation ?target .}};
> SPARQL_TO_SQL_TEXT
> LONG VARCHAR
> _______________________________________________________________________________
>
> SELECT TOP 1 1 AS __ask_retval
> FROM DB.DBA.RDF_QUAD AS "s_1_2_t0"
> INNER JOIN DB.DBA.RDF_QUAD AS "s_1_2_t1"
> ON ( /* two fields belong to same equiv */
> /* retval[ */ "s_1_2_t1"."S" /* subject */ /* ]retval */ = /*
> retval[ */ "s_1_2_t0"."S" /* subject */ /* ]retval */)
> WHERE /* field equal to URI ref */
> "s_1_2_t0"."G" = __i2idn ( /* UNAME as sqlval */ __bft (
> 'http://sparq.uniprot.org/uniprot' <http://sparq.uniprot.org/uniprot'> , 1))
> AND /* field equal to URI ref */
> "s_1_2_t0"."P" = __i2idn ( /* UNAME as sqlval */ __bft (
> 'http://www.w3.org/1999/02/22-rdf-syntax-ns#type'
> <http://www.w3.org/1999/02/22-rdf-syntax-ns#type'> , 1))
> AND /* field equal to URI ref */
> "s_1_2_t0"."O" = __i2idn ( /* UNAME as sqlval */ __bft (
> 'http://purl.uniprot.org/core/Protein'
> <http://purl.uniprot.org/core/Protein'> , 1))
> AND /* field equal to URI ref */
> "s_1_2_t1"."G" = __i2idn ( /* UNAME as sqlval */ __bft (
> 'http://sparq.uniprot.org/uniprot' <http://sparq.uniprot.org/uniprot'> , 1))
> AND /* field equal to URI ref */
> "s_1_2_t1"."P" = __i2idn ( /* UNAME as sqlval */ __bft (
> 'http://purl.uniprot.org/core/annotation'
> <http://purl.uniprot.org/core/annotation'> , 1))
>
> However, to be honest. This is not a standard setup of the quad table.
>
> It happens if adding this index
> CREATE BITMAP INDEX RDF_QUAD_P ON DB.DBA.RDF_QUAD (P)
> or
> CREATE DISTINCT NO PRIMARY KEY REF BITMAP INDEX RDF_QUAD_GP ON RDF_QUAD
> (G, P) PARTITION (P INT (0hexffff00))
>
> and then running the query.
> But funnily if you then restart the db, it works again!
>
> Regards,
> Jerven
>
>
>
> On 2017-02-09 17:28, Hugh Williams wrote:
>> Hi Jerven,
>>
>> Thanks for the update, glad to hear you found the cause of the problem
>> …
>>
>> Best Regards
>> Hugh Williams
>> Professional Services
>>
>> OpenLink Software, Inc. //
>> http://www.openlinksw.com/ <http://www.openlinksw.com/>
>> Weblog -- http://www.openlinksw.com/blogs/
>> <http://www.openlinksw.com/blogs/>
>> LinkedIn -- http://www.linkedin.com/company/openlink-software/
>> <http://www.linkedin.com/company/openlink-software/>
>> Twitter -- http://twitter.com/OpenLink <http://twitter.com/OpenLink>
>> Google+ -- MAILSCANNER HAS DETECTED DEFINITE FRAUD IN THE WEBSITE AT
>> "PLUS.GOOGLE.COM <http://plus.google.com/>". DO _NOT_ TRUST THIS WEBSITE:
>> http://plus.google.com/100570109519069333827/
>> <http://plus.google.com/100570109519069333827/> [2]
>> Facebook -- http://www.facebook.com/OpenLinkSoftware
>> <http://www.facebook.com/OpenLinkSoftware>
>> Universal Data Access, Integration, and Management Technology
>> Providers
>>
>>> On 9 Feb 2017, at 13:35, Jerven Tjalling Bolleman
>>> <Jerven.Bolleman@sib.swiss <mailto:Jerven.Bolleman@sib.swiss>> wrote:
>>> Hi Hugh,
>>>
>>> The source of the error was a truncated database file on my
>>> developer desktop due to disk quota and a problem between my office
>>> chair and my desk ;)
>>>
>>> Regards,
>>> Jerven
>>>
>>> On 2017-02-06 20:58, Hugh Williams wrote:
>>> Hi Jerven,
>>> Does the same error occur when the query is run against the /sparql
>>> endpoint or isql ie not via the JDBC Driver ?
>>> Also, are you able to get the query “profile” or
>>> “sparql_to_sql_text” for the query as detailed at:
>>>
>> https://virtuoso.openlinksw.com/dataspace/doc/dav/wiki/Main/VirtTipsAndTricksAanalyzingSPARQLQuery
>>
>> <https://virtuoso.openlinksw.com/dataspace/doc/dav/wiki/Main/VirtTipsAndTricksAanalyzingSPARQLQuery>
>>> What version of Virtuoso is being used, is it a latest 07.20.3217
>>> build ?
>>> Best Regards
>>> Hugh Williams
>>> Professional Services
>>> OpenLink Software, Inc. //
>>> http://www.openlinksw.com/
>>> Weblog -- http://www.openlinksw.com/blogs/
>>> LinkedIn -- http://www.linkedin.com/company/openlink-software/
>>> Twitter -- http://twitter.com/OpenLink
>>> Google+ -- MAILSCANNER HAS DETECTED DEFINITE FRAUD IN THE WEBSITE
>>> AT
>>> "MAILSCANNER HAS DETECTED DEFINITE FRAUD IN THE WEBSITE AT
>>> "PLUS.GOOGLE.COM". DO _NOT_ TRUST THIS WEBSITE: PLUS.GOOGLE.COM
>>> [1]". DO _NOT_ TRUST THIS WEBSITE:
>>> MAILSCANNER HAS DETECTED DEFINITE FRAUD IN THE WEBSITE AT
>>> "PLUS.GOOGLE.COM". DO _NOT_ TRUST THIS WEBSITE:
>>> http://plus.google.com/100570109519069333827/ [2] [2]
>>> Facebook -- http://www.facebook.com/OpenLinkSoftware
>>> Universal Data Access, Integration, and Management Technology
>>> Providers
>>> On 6 Feb 2017, at 16:08, Jerven Tjalling Bolleman
>>> <jerven.bolleman@sib.swiss> wrote:
>>> Dear Virtuoso users,
>>> I am getting an new error from virtuoso, during development of a new
>>> feature
>>> SPARQL execute failed:[SELECT (COUNT(?target) AS ?targets) { GRAPH
>>> <http://sparql.uniprot.org/uniprot>{?subject a
>>> <http://purl.uniprot.org/core/MRNA> . ?target a
>>> <http://biohackathon.org/resource/faldo#ExactPosition> . ?subject
>>> <http://purl.uniprot.org/core/author> ?target .}}]
>>> Exception:virtuoso.jdbc4.VirtuosoException: SQ156: Internal
>>> Optimized
>>> compiler error : sqlo table has no index in sqldf.c:3777.
>>> Please report the statement compiled.
>>> Regards,
>>> Jerven
>>> --
>>> Jerven Tjalling Bolleman
>>> SIB | Swiss Institute of Bioinformatics
>>> CMU - 1, rue Michel Servet - 1211 Geneva 4
>>> t: +41 22 379 58 85 - f: +41 22 379 58 58
>>> Jerven.Bolleman@sib.swiss - http://www.sib.swiss [3]
>>>
>> ------------------------------------------------------------------------------
>>> Check out the vibrant tech community on one of the world's most
>>> engaging tech sites, SlashDot.org [4] [1]! http://sdm.link/slashdot
>>> _______________________________________________
>>> Virtuoso-users mailing list
>>> Virtuoso-users@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/virtuoso-users
>>> Links:
>>> ------
>>> [1] http://slashdot.org [4]
>>> [2] MAILSCANNER HAS DETECTED DEFINITE FRAUD IN THE WEBSITE AT
>>> "PLUS.GOOGLE.COM". DO _NOT_ TRUST THIS WEBSITE:
>>> http://plus.google.com/100570109519069333827/ [2]
>>>
>> ------------------------------------------------------------------------------
>>> Check out the vibrant tech community on one of the world's most
>>> engaging tech sites, SlashDot.org [4]! http://sdm.link/slashdot
>>> _______________________________________________
>>> Virtuoso-users mailing list
>>> Virtuoso-users@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/virtuoso-users
>>
>> --
>> Jerven Tjalling Bolleman
>> SIB | Swiss Institute of Bioinformatics
>> CMU - 1, rue Michel Servet - 1211 Geneva 4
>> t: +41 22 379 58 85 - f: +41 22 379 58 58
>> Jerven.Bolleman@sib.swiss - http://www.sib.swiss [3]
>>
>>
>> Links:
>> ------
>> [1] http://plus.google.com/
>> [2] http://plus.google.com/100570109519069333827/
>> [3] http://www.sib.swiss/
>> [4] http://slashdot.org/
>
> --
> Jerven Tjalling Bolleman
> SIB | Swiss Institute of Bioinformatics
> CMU - 1, rue Michel Servet - 1211 Geneva 4
> t: +41 22 379 58 85 - f: +41 22 379 58 58
> Jerven.Bolleman@sib.swiss <mailto:Jerven.Bolleman@sib.swiss> -
> http://www.sib.swiss <http://www.sib.swiss/>
>
>
> ------------------------------------------------------------------------------
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, SlashDot.org <http://slashdot.org/>!
> http://sdm.link/slashdot <http://sdm.link/slashdot>
> _______________________________________________
> Virtuoso-users mailing list
> Virtuoso-users@lists.sourceforge.net
> <mailto:Virtuoso-users@lists.sourceforge.net>
> https://lists.sourceforge.net/lists/listinfo/virtuoso-users
> <https://lists.sourceforge.net/lists/listinfo/virtuoso-users>
smime.p7s
Description: S/MIME cryptographic signature
------------------------------------------------------------------------------ Check out the vibrant tech community on one of the world's most engaging tech sites, SlashDot.org! http://sdm.link/slashdot
_______________________________________________ Virtuoso-users mailing list Virtuoso-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/virtuoso-users
