Hi Andy, thanks for your quick reply!
No, I did not configure timeouts anywhere else. I created the TDB datasets using the Fuseki web interface and then used tdbloader2 to upload the triples into their respective TDB stores (with a Bash script to make life easier). Where else can I set this timeout option? Do I have to modify run/config.ttl? Just guessing - correct me if I’m wrong: It seems to me that the TDB store itself is configured in RFD. Duh - this makes life much more complicated! Where can I find more information about this? Is this specific to Fuseki or TDB? Sebastian Am 17.04.2015 um 19:48 schrieb Andy Seaborne <[email protected]>: > Hi Sebastian, > > Are timeout also set anywhere else such as the service/dataset configuration? > If so, they will override the global (server-wide) default. > > The 503 message is confusing - it prints the gloabl default (100000ms) even > if the query executed under a different timeout. It looks like it it set to > 3000ms somewhere else as well. > > Andy > > On 17/04/15 13:49, Mate, Sebastian wrote: >> Dear Fuseki users, >> >> I want to use Fuseki 2 for querying various big ontologies, e.g. dbpedia. >> However, I have encountered the problem that Fuseki is ignoring the timeout >> setting. >> >> I started the server with >> >> ./fuseki-server --timeout=100000 >> >> which should set the timeout to 100s, but it is rather 3s (see below). I >> have seen that there were some discussions regarding the handling of >> timeouts, but I'm not sure if this is related to my problem. Here's the log >> - please note that I'm using one of the latest snapshot builds (the official >> 2.0.0 release is behaving the same way for me): >> >> [2015-04-17 09:53:14] Server INFO Fuseki 2.0.1-SNAPSHOT >> 2015-04-16T16:45:46+0000 >> ... >> [2015-04-17 09:53:58] Config INFO Register: >> /dbpedia_topical_concepts_unredirected_en >> [2015-04-17 09:53:58] Config INFO Register: /dbpedia_wikipedia_links_en >> [2015-04-17 09:53:58] Server INFO Started 2015/04/17 09:53:58 MESZ on >> port 3030 >> [2015-04-17 09:54:12] Fuseki INFO [1] POST >> http://xxx.xxx.xxx.xxx:3030/dbpedia_page_links_unredirected_en/query >> [2015-04-17 09:54:12] Fuseki INFO [1] POST >> /dbpedia_page_links_unredirected_en :: 'query' :: >> [application/x-www-form-urlencoded charset=UTF-8] ? >> [2015-04-17 09:54:12] Fuseki INFO [1] Query = select (count(*) as >> ?count) {?s ?p ?o} >> [2015-04-17 09:54:15] Fuseki INFO [1] 503 The query timed out >> (restricted to 100000 ms) (3,098 s) >> >> The problem appears with other queries as well. Sometimes limiting the >> result set (e.g. by appending "LIMIT 20" to the SPARQL query) helps. >> >> The system is a quad core virtual machine with 32 GB RAM, running Ubuntu >> 14.04. "java -version" is reporting: >> >> OpenJDK Runtime Environment (IcedTea 2.5.4) (7u75-2.5.4-1~trusty1) >> OpenJDK 64-Bit Server VM (build 24.75-b04, mixed mode) >> >> I would highly appreciate any advice. I really like Fuseki and would love to >> use it for some data mining in medical informatics. >> >> Many thanks, >> Sebastian >> >> ------------------------------------------------------------------------------------------ >> Dipl.-Inf. Sebastian Mate >> >> Lehrstuhl für Medizinische Informatik >> Friedrich-Alexander-Universität Erlangen-Nürnberg >> Wetterkreuz 13, 91058 Erlangen-Tennenlohe, Germany >> >> Tel. +49 (9131) 85-26722 // Fax +49 (9131) 85-26754 >> eMail: [email protected] >> >> >> >> >> >
