I tried a number of different JDKs: Oracle JDK 6u45, Oracle JDK 7 u45 and u51, OpenJDK 7.
My system at home has a i7-2770k, and I’ve tried with both 4 & 8 CPUs assigned to the VM. I’ll try increasing the Thread.sleep(200) at the end when I get home tonight. Ron On Mar 3, 2014, at 9:34 AM, Seth Leger <s...@opennms.org> wrote: > Hi Ron, > > What JVM are you using where you are encountering this issue? The test > spawns 2000 threads but only about 10 should be running at any time. > However, on a slower machine than my development machine, maybe you > could get problems if the threads aren't completing before the next set > of 10 spawns. Can you edit the test and try increasing the > Thread.sleep(200) at the end until your system works? Or if you want, > you could change the test to use a set of thread pools. The idea is just > to have several threads all running at the same time, connecting to the > database with contention, and throwing SQLExceptions. > > This test exists because this case was causing problems with the > Atomikos connection pool when running outside of a JTA transaction > manager until I made a change to the test query for the pool. > > -- Seth > > > On 3/2/14, 11:24 PM, Ronald Roskens wrote: >> I’m having trouble running the unit tests for master right now. Everything >> is running ok until it starts working on org.opennms.core.db and then I get >> an OutOfMemory exception that fails the test run there. I’m using the >> default maven opts. >> >> If I change the ConnectionFactoryTest so it only creates 200 threads instead >> of 2000, the unit tests run through to completion for all projects. >> >> My VM has 10GB of memory, but I’m also running instances of jenkins, sonar, >> and nexus in it so I don’t have too much extra memory to dedicate towards >> these tests. >> >> Would making the unit test take a property for the number of threads be >> acceptable or is there a better way of running that test? >> >> Ron >> >> >> >> >> ------------------------------------------------------------------------------ >> Subversion Kills Productivity. Get off Subversion & Make the Move to >> Perforce. >> With Perforce, you get hassle-free workflows. Merge that actually works. >> Faster operations. Version large binaries. Built-in WAN optimization and the >> freedom to use Git, Perforce or both. Make the move to Perforce. >> http://pubads.g.doubleclick.net/gampad/clk?id=122218951&iu=/4140/ostg.clktrk >> _______________________________________________ >> Please read the OpenNMS Mailing List FAQ: >> http://www.opennms.org/index.php/Mailing_List_FAQ >> >> opennms-devel mailing list >> >> To *unsubscribe* or change your subscription options, see the bottom of this >> page: >> https://lists.sourceforge.net/lists/listinfo/opennms-devel > > > ------------------------------------------------------------------------------ > Subversion Kills Productivity. Get off Subversion & Make the Move to Perforce. > With Perforce, you get hassle-free workflows. Merge that actually works. > Faster operations. Version large binaries. Built-in WAN optimization and the > freedom to use Git, Perforce or both. Make the move to Perforce. > http://pubads.g.doubleclick.net/gampad/clk?id=122218951&iu=/4140/ostg.clktrk > _______________________________________________ > Please read the OpenNMS Mailing List FAQ: > http://www.opennms.org/index.php/Mailing_List_FAQ > > opennms-devel mailing list > > To *unsubscribe* or change your subscription options, see the bottom of this > page: > https://lists.sourceforge.net/lists/listinfo/opennms-devel ------------------------------------------------------------------------------ Subversion Kills Productivity. Get off Subversion & Make the Move to Perforce. With Perforce, you get hassle-free workflows. Merge that actually works. Faster operations. Version large binaries. Built-in WAN optimization and the freedom to use Git, Perforce or both. Make the move to Perforce. http://pubads.g.doubleclick.net/gampad/clk?id=122218951&iu=/4140/ostg.clktrk _______________________________________________ Please read the OpenNMS Mailing List FAQ: http://www.opennms.org/index.php/Mailing_List_FAQ opennms-devel mailing list To *unsubscribe* or change your subscription options, see the bottom of this page: https://lists.sourceforge.net/lists/listinfo/opennms-devel
