3

I have graphDB 8.0.2 running on an Ubuntu server. I run it with /var/lib/graphdb/bin/graphdb -Xmx1224M -Xms712M &. The graphdb process stops with no reason. I don't see any errors in the logs (see below). What could be the reason for this ? How can I debug this ? I already increased the memory given to the process, but still experiments this behavior. Would even more memory solve anything ?

Thank you for your help

[INFO ] 2017-09-06 13:38:58,092 [repositories/sem-bo | o.e.r.h.s.r.TupleQueryResultView] Request for query -1311463370 is finished [INFO ] 2017-09-06 13:38:58,097 [repositories/sem-bo | c.o.f.s.RepositoryController] POST query -2012703728 [INFO ] 2017-09-06 13:38:58,114 [repositories/sem-bo | o.e.r.h.s.r.TupleQueryResultView] Request for query -2012703728 is finished [INFO ] 2017-09-06 13:43:43,796 [Thread-4 | c.o.g.s.GraphDB] Stopping GraphDB. [INFO ] 2017-09-06 13:43:43,964 [localhost-startStop-2 | c.o.f.c.s.SemanticDataManagement] Shutting down current semantic location [INFO ] 2017-09-06 13:43:44,381 [Thread-5 | c.o.t.s.i.PluginManager] Shutting down plugins (DEFAULT)... [INFO ] 2017-09-06 13:43:44,415 [Thread-5 | c.o.t.b.AVLRepository] NumberOfStatements = 45247 [INFO ] 2017-09-06 13:43:44,415 [Thread-5 | c.o.t.b.AVLRepository] NumberOfExplicitStatements = 45089 [INFO ] 2017-09-06 13:43:44,415 [Thread-5 | c.o.t.f.GraphDBFreeSchemaRepository] Shutting down entity pool [INFO ] 2017-09-06 13:43:49,444 [Thread-5 | c.o.t.f.GraphDBFreeSchemaRepository] Entity pool was shut down

ThomasFrancart
  • 470
  • 3
  • 13
  • 2
    I don't know the GraphDB startup scripts too well, but the logs suggest that it is doing a "normal" shutdown. You're starting as a background task (`&`) from a shell. Do you at some point close that shell - because that will send an exit signal to any running background task connected to it. – Jeen Broekstra Sep 07 '17 at 02:00
  • Yes this was probably my stupid mistake. I thought adding "&" would keep the process alive when the shell terminates, but this is not the case. Thanks Jeen – ThomasFrancart Sep 08 '17 at 07:24

1 Answers1

3

Based on the logs the only thing that comes to mind is something sends the interrupt signal to GraphDB and it's quitting gracefully.

The recommended way to run GraphDB in the background is by adding the -d (daemonize) option, e.g. in your case /var/lib/graphdb/bin/graphdb -Xmx1224M -Xms712M -d.

Pavel Mihaylov
  • 351
  • 1
  • 3