On Wed, Dec 22, 2010 at 6:11 PM, Marko Rodriguez <okramma...@gmail.com> wrote:
> So I bumped Blueprints 0 .4-SNAPSHOT to use 1.2M06 and during the test suite > evaluation, I was getting fatty stack traces being thrown all up in my face > left, right, and yes, center. As long as they are not coming from below, there is no reason for alarm. However... > My random thought on the matter---Luca and I have been having *potentially* > similar problems with OrientDB and shutting down, opening, and deleting the > DB really fast. Don't > know if this is the same issue now showing up in > Neo4j? The TestSuite will, in general, create a graph (directory), use it, > close it, open it, delete the directory, create a > new graph (directory), > etc. Is the OS not having happy good time? Why happy good time with 1.2M05? I slapped up a loop doing pretty much that - create a db, store+index a few things, close, [open, read, close] (in a loop), delete, do over, for around 50 iterations. Using straight up (no tinkerpop stack) vanilla 1.2M06 on Ubuntu 10.10. No problems so far. I will leave it to run for a while, to teach my HDD some manners and I will check it again later. That lock file left behind, however, smells like crash/improper shutdown (or overlapping instances?). That really fast open/shutdown cycle you speak of, how do you do it, shell script or from main()? Does the problem appear in every cycle or intermittently? cheers, CG _______________________________________________ Neo4j mailing list User@lists.neo4j.org https://lists.neo4j.org/mailman/listinfo/user
