From: Kevin F. <kf...@ar...> - 2017-12-13 19:08:52
|
What's in the tmp directory? Look, too, for a growing log file. Those two seem like logical places to start, so apologies if you've already covered these locales. Yours, Kevin On 12/12/17 02:33, Rune Stilling wrote: > If someone takes a look this at some point it now seems that the import process has stalled. Since yesterday the invisible disk usage has gone up, but the datafile hasn’t been touched. > > Yesterday: > -rw-r--r-- 1 root root 186G 11 dec 11:46 blazegraph.jnl > /dev/xvda1 2,0T 1,7T 344G 83% / > > Today: > -rw-r--r-- 1 root root 186G 11 dec 11:46 blazegraph.jnl > /dev/xvda1 2,0T 1,9T 73G 97% / > > Soon the importer will probably stop with an out of disk space exception. > > /Rune > >> Den 11. dec. 2017 kl. 10.24 skrev Rune Stilling <rs...@un...>: >> >> Hi list >> >> An update on my question. It seems that the blazegraph importer makes use of disk space besides the datafile. Currently I’m running the import on a server with 2 TB of disk space. Even though the blazegraph.jnl file is only around 186 GB (and still running the import) the disk reports a use of 1.7 TB. I can’t find any files that make up for this use though. >> >> If the import doesn’t finish soon the import will end with the same exception again. >> >> Is this normal behavior and what (hidden) files is the blazegraph importer creating during the import? >> >> Reagrds, >> Rune >> >> ~]$ ls -la ~/blazegraph --block-size=G >> totalt 171G >> drwxrwxr-x 2 ec2-user ec2-user 1G 4 dec 12:25 . >> drwx------ 6 ec2-user ec2-user 1G 29 nov 10:52 .. >> -rw-r--r-- 1 root root 186G 11 dec 09:21 blazegraph.jnl >> >> [ ~]$ df -h >> Filsystem Størr Brugt Tilb Brug% Monteret på >> devtmpfs 15G 60K 15G 1% /dev >> tmpfs 15G 0 15G 0% /dev/shm >> /dev/xvda1 2,0T 1,7T 363G 83% / >> >>> Den 4. dec. 2017 kl. 07.45 skrev Rune Stilling <rs...@un...>: >>> >>> Hi list >>> >>> I have setup an AWS instance using the blazegraph-ami-2.1.5. I have done the setup using a one disc configuration with 1 terabyte of space. After this I use the biodata.jar to import the full Wikidata dataset: >>> >>>> sudo java -cp bigdata.jar com.bigdata.rdf.store.DataLoader -namespace wikidata -defaultGraph http://www.wikidata.org fullfeat >>>> ure.properties latest-all.ttl >>> >>> >>> After running 4 days the import terminates with an exception saying “no more space left on device”. There’s plenty of space left so what can I do to make this work? >>> >>>> [ec2-user@ip-10-10-0-189 ~]$ df -h >>>> Filsystem Størr Brugt Tilb Brug% Monteret på >>>> devtmpfs 15G 60K 15G 1% /dev >>>> tmpfs 15G 0 15G 0% /dev/shm >>>> /dev/xvda1 1008G 305G 703G 31% / >>> >>> >>>> [ec2-user@ip-10-10-0-189 ~]$ ls -la ~/blazegraph --block-size=GB >>>> totalt 119GB >>>> drwxrwxr-x 2 ec2-user ec2-user 1GB 29 nov 10:53 . >>>> drwx------ 6 ec2-user ec2-user 1GB 29 nov 10:52 .. >>>> -rw-r--r-- 1 root root 124GB 2 dec 10:14 blazegraph.jnl >>> >>> >>> /Rune Stilling >>> >> > > > ------------------------------------------------------------------------------ > Check out the vibrant tech community on one of the world's most > engaging tech sites, Slashdot.org! http://sdm.link/slashdot > _______________________________________________ > Bigdata-developers mailing list > Big...@li... > https://lists.sourceforge.net/lists/listinfo/bigdata-developers > |