Hi, I've added a sanity check to verify weather all the CacheManagers created by a test are being shutdown within the same test[1]. The cool part is that if a CM is not being closed, the fwk will report you where the unclosed CacheManager was created (java file and line, see bellow), so that you don't have to spend too much time looking into code ;) E.g. [pool-2-thread-7] Test testTxChangesOnAtomicMap(org.infinispan.atomic.AtomicMapFunctionalTest) succeeded. Test suite progress: tests succeeded: 179, failed: 1, skipped: 0. [pool-2-thread-7] Test testTxChangesOnAtomicMapNoLocks(org.infinispan.atomic.AtomicMapFunctionalTest) succeeded. Test suite progress: tests succeeded: 180, failed: 1, skipped: 0. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!! (pool-2-thread-7) Exiting because atomic.AtomicMapFunctionalTest has NOT shut down all the cache managers it has started !!!!!!! !!!!!! (pool-2-thread-7) The still-running cacheManager was created here: org.infinispan.atomic.AtomicMapFunctionalTest.setUp(AtomicMapFunctionalTest.java:33) !!!!!!! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! Cheers, Mircea [1]https://jira.jboss.org/jira/browse/ISPN-314 _______________________________________________ infinispan-dev mailing list infinispan-dev(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/infinispan-dev

Thanks! I've been using testNG listener to notify me whenever a thread finishes processing a test. Whenever a cache is created, I create a stack trace and determine which method performed the call: private static String extractMethodName() { StackTraceElement[] stack = Thread.currentThread().getStackTrace(); if (stack.length == 0) return null; for (int i = stack.length - 1; i > 0; i--) { StackTraceElement e = stack[i]; String className = e.getClassName(); if ((className.indexOf("org.infinispan") != -1) && className.indexOf("org.infinispan.test") < 0) return e.toString(); } return null; } On 24 Feb 2010, at 18:12, Vladimir Blagojevic wrote: > Very cool! You are even reporting the line number! How did you do this > CacheManager tracking? > On 2010-02-24, at 11:06 AM, Mircea Markus wrote: > >> Hi, >> >> I've added a sanity check to verify weather all the CacheManagers >> created by a test are being shutdown within the same test[1]. >> The cool part is that if a CM is not being closed, the fwk will report >> you where the unclosed CacheManager was created (java file and line, >> see bellow), so that you don't have to spend too much time looking >> into code ;) >> E.g. >> >> [pool-2-thread-7] Test >> testTxChangesOnAtomicMap(org.infinispan.atomic.AtomicMapFunctionalTest) >> succeeded. >> Test suite progress: tests succeeded: 179, failed: 1, skipped: 0. >> [pool-2-thread-7] Test >> testTxChangesOnAtomicMapNoLocks(org.infinispan.atomic.AtomicMapFunctionalTest) >> succeeded. >> Test suite progress: tests succeeded: 180, failed: 1, skipped: 0. >> >> !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! >> !!!!!! (pool-2-thread-7) Exiting because >> atomic.AtomicMapFunctionalTest has NOT shut down all the cache >> managers it has started !!!!!!! >> !!!!!! (pool-2-thread-7) The still-running cacheManager was created >> here: >> org.infinispan.atomic.AtomicMapFunctionalTest.setUp(AtomicMapFunctionalTest.java:33) >> !!!!!!! >> !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! >> >> Cheers, >> Mircea >> >> [1]https://jira.jboss.org/jira/browse/ISPN-314 >> _______________________________________________ >> infinispan-dev mailing list >> infinispan-dev(a)lists.jboss.org >> https://lists.jboss.org/mailman/listinfo/infinispan-dev > > > _______________________________________________ > infinispan-dev mailing list > infinispan-dev(a)lists.jboss.org > https://lists.jboss.org/mailman/listinfo/infinispan-dev _______________________________________________ infinispan-dev mailing list infinispan-dev(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/infinispan-dev