Hi Jakob, all To verify that this issue only affects H2 (and not Postgres) I configured a Postgres and run the tests using it. With this change the test run fine for both a clean and subsequent builds.
As the reported bugs seam to be limited to H2 running on Mac and H2 is anyway only intended to be used in demo installation I see no reason to cancel the release because of that. While testing with Postgres I also found an other Unit Test issue when using postgres with the LDCache KiWi Backend (reported with MARMOTTA-490[1]) So with MARMOTTA-489 and MARMOTTA-490 reported as affecting 3.2.0 a +1 for the release from my side. best Rupert [1] https://issues.apache.org/jira/browse/MARMOTTA-490 On Mon, Apr 14, 2014 at 7:22 PM, Thomas Kurz <[email protected]> wrote: > Jakob, > > for me it is okay to release, because a) other systems than mac os do not > have problems to build, b) it is not clear if the bug appears outside the > tests and the h2 and c) if we wait for years, the release will also not be > bugfree ;) > > So, +1 for the release. But we have to be aware of getting some emails from > surprised users! > > Thomas > >> Am 14.04.2014 um 16:57 schrieb Jakob Frank <[email protected]>: >> >> Rupert, Thomas, >> >> thanks for checking! This is indeed something we need to look into. >> Could you please file an Issue in JIRA? >> >> Now what does this mean for the release vote? Do you "-1"? >> >> @all: WDYT - Should we delay the release (again) until the issue is >> fixed or is it OK to continue? >> >> Thanks, >> Jakob >> >>> On 14 April 2014 16:43, Thomas Kurz <[email protected]> wrote: >>> Hello! >>> >>> Second try ;) : >>> >>> For me marmotta is not building, too. And also not always with the same >>> error message. The last on I got was: >>> --- >>> >>> Failed tests: >>> >>> HazelcastClusterTest>BaseClusterTest.testClusteredCacheBNode:113 null >>> >>> HazelcastClusterTest>BaseClusterTest.testClusteredCacheUri:85 null >>> >>> >>> Tests in error: >>> >>> HazelcastClusterTest>BaseClusterTest.testRegistry:155 » NullPointer >>> >>> >>> Tests run: 208, Failures: 2, Errors: 1, Skipped: 133 >>> >>> --- >>> >>> I am on MAC, too. java version "1.7.0_51". The first build failed with a >>> concurrent database operation failure. >>> >>> Best regards >>> >>> Thomas >>> >>> >>> On Mon, Apr 14, 2014 at 4:25 PM, Rupert Westenthaler < >>> [email protected]> wrote: >>> >>>> Hi all, >>>> >>>> I had problems with some unit tests related to concurrency of the KiWi >>>> Triplestore on mac. As the first build failed I made around 20 >>>> additional builds with diver gating results. I also wiped the local >>>> repository twice to simulate several clean builds: >>>> >>>> >>>> On every first build I got an error in the AfterClass of the H2 >>>> concurrency Test (kiwi-triplestore module). >>>> >>>> org.apache.marmotta.kiwi.test.H2ConcurrencyTest Time elapsed: >>>> 3.64 sec <<< FAILURE! >>>> java.lang.AssertionError: null >>>> [..] >>>> at >>>> org.apache.marmotta.kiwi.test.H2ConcurrencyTest.dropDatabase(H2ConcurrencyTest.java:63) >>>> >>>> Follow up builds (with -rf :kiwi-triplestore) have never shown this. >>>> New build sometimes do. Looks like if this is somehow related on how >>>> much Disc IO was done in-between two tests ... >>>> >>>> I also randomly got failures that do look like they where all related >>>> to concurrency (kiwi-triplestore module) (in –20% of the tests) >>>> >>>> >>>> testPreparedTupleQuery2(org.apache.marmotta.kiwi.test.sesame.repository.KiWiRepositoryConnectionTest) >>>> Time elapsed: 0.011 sec <<< FAILURE! >>>> java.lang.AssertionError: unexpected value for name: "Alice" >>>> expected:<"Bob"> but was:<"Alice"> >>>> [..] >>>> at >>>> org.apache.marmotta.kiwi.test.junit.KiWiDatabaseRunner$DatabaseTestClassRunner$ExecutionLogger$1.evaluate(KiWiDatabaseRunner.java:313) >>>> at >>>> org.apache.marmotta.kiwi.test.junit.KiWiDatabaseRunner$DatabaseTestClassRunner$CheckDBRule$1.evaluate(KiWiDatabaseRunner.java:295) >>>> at >>>> org.junit.internal.runners.statements.FailOnTimeout$StatementThread.run(FailOnTimeout.java:74) >>>> >>>> There was also (once) an issue in the reasoner - could be also related >>>> to concurrency as AFAIK the reasoning is also done in an own thread >>>> >>>> >>>> testResolveBaseTriplesMulti(org.apache.marmotta.kiwi.reasoner.test.engine.JustificationResolutionTest) >>>> Time elapsed: 0.216 sec <<< FAILURE! >>>> java.lang.AssertionError: expected:<2> but was:<4> >>>> [..] >>>> at >>>> org.apache.marmotta.kiwi.reasoner.test.engine.JustificationResolutionTest.testResolveBaseTriplesMulti(JustificationResolutionTest.java:190) >>>> >>>> Finally two times I got what seams to be a deadlock of some sort in >>>> the org.apache.marmotta.platform.core.test.config.ConfigurationServiceTest >>>> >>>> Thread qtp614303200-48 - /marmotta/resource/YR6ENRVc is waiting on >>>> an active Exclusive lock acquired in qtp614303200-48 - >>>> /marmotta/resource/YR6ENRVc >>>> java.lang.Throwable: null >>>> at >>>> info.aduna.concurrent.locks.LockManager.logStalledLock(LockManager.java:247) >>>> [sesame-sail-api-2.7.11.jar:na] >>>> [...] >>>> at >>>> org.openrdf.repository.base.RepositoryConnectionWrapper.setNamespace(RepositoryConnectionWrapper.java:489) >>>> [sesame-repository-api-2.7.11.jar:na] >>>> at >>>> org.apache.marmotta.platform.core.services.triplestore.ContextServiceImpl.checkConnectionNamespace(ContextServiceImpl.java:150) >>>> [classes/:na] >>>> >>>> This exception was repeated every some minutes. I both cases I stopped >>>> the process after ~10min. >>>> >>>> Pattform: >>>> >>>> java -version >>>> java version "1.7.0_45" >>>> Java(TM) SE Runtime Environment (build 1.7.0_45-b18) >>>> Java HotSpot(TM) 64-Bit Server VM (build 24.45-b08, mixed mode) >>>> >>>> OSX 10.9.2 on MacBook Pro Retina, Mid 2012 >>>> >>>> best >>>> Rupert >>>> >>>> >>>>> On Mon, Apr 14, 2014 at 2:14 PM, Jakob Frank <[email protected]> wrote: >>>>> Sorry, my +1 did not go to the list... >>>>> >>>>> Best, >>>>> Jakob >>>>> >>>>>> On 9 April 2014 14:19, Jakob Frank <[email protected]> wrote: >>>>>> And of course, there's my +1 ;-) >>>>>> >>>>>> Best >>>>>> Jakob >>>>>> >>>>>>> On 9 April 2014 14:19, Jakob Frank <[email protected]> wrote: >>>>>>> Dear Marmotta community, >>>>>>> >>>>>>> finally our first release as a top-level project is about to happen, it >>>>>>> time to VOTE! >>>>>>> >>>>>>> A candidate for the Marmotta 3.2.0 release is available at: >>>>>>> >>>>>>> https://dist.apache.org/repos/dist/dev/marmotta/3.2.0/ >>>>>>> >>>>>>> The release candidate is based on the sources tagged with 3.2.0 in: >>>>>>> >>>>>>> https://git-wip-us.apache.org/repos/asf/marmotta.git >>>>>>> >>>>>>> The checksum of this commit is f5b5cf0c21ab3ce0aaea1e816658e0aa2d59bb86 >>>>>>> >>>>>>> The release candidate consists of the following source distribution >>>>>>> archives: >>>>>>> - apache-marmotta-3.2.0-src.[zip|tar.gz] >>>>>>> SHA1 of ZIP: cb1505f7de914cb37c6545d5d0727440d850555b >>>>>>> SHA1 of TGZ: 25482e2fb39124895e63e33ebec6712af31c9e53 >>>>>>> >>>>>>> In addition, the following supplementary binary distributions are >>>>>>> provided: >>>>>>> - apache-marmotta-3.2.0-installer.[zip|tar.gz] >>>>>>> SHA1 of ZIP: 12151c2f25561aeff1667bd97b5ef199a5acf6cc >>>>>>> SHA1 of TGZ: 053b8d96121bb7e542207ebdab2fa4bd6fa2369f >>>>>>> - apache-marmotta-3.2.0-ldpath.[zip|tar.gz] >>>>>>> SHA1 of ZIP: b4e5a784d17fd55f066554992f97f39731567aec >>>>>>> SHA1 of TGZ: 731dfc6423cbe3d82dd851cee75a8b9a44667e86 >>>>>>> - apache-marmotta-3.2.0-webapp.[zip|tar.gz] >>>>>>> SHA1 of ZIP: ad8469c60a69dcfe446f279d0f9e9578b76ce19b >>>>>>> SHA1 of TGZ: 40761cc7ef32a1e9643767dd306d886ae2981073 >>>>>>> >>>>>>> A staged Maven repository is available for review at: >>>>>>> >>>>>>> >>>> https://repository.apache.org/content/repositories/orgapachemarmotta-1008 >>>>>>> >>>>>>> The changelog for the 3.2.0 version is available from JIRA: >>>>>>> >>>>>>> >>>> https://issues.apache.org/jira/secure/ReleaseNote.jspa?projectId=12314321&version=12324278 >>>>>>> >>>>>>> >>>>>>> Please vote on releasing this package as Apache Marmotta 3.2.0. >>>>>>> The vote is open for at least 72 hours and passes if a majority of at >>>>>>> least three +1 Marmotta PMC votes are cast. >>>>>>> >>>>>>> [ ] +1 Release this package as Apache Marmotta 3.2.0 >>>>>>> [ ] 0 I don't feel strongly about it, but I'm okay with the release >>>>>>> [ ] -1 Do not release this package because... >>>>>>> >>>> >>>> >>>> >>>> -- >>>> | Rupert Westenthaler [email protected] >>>> | Bodenlehenstraße 11 ++43-699-11108907 >>>> | A-5500 Bischofshofen >>>> | >>>> REDLINK.CO.......................................................................... >>>> | http://redlink.co/ >>>> -- | Rupert Westenthaler [email protected] | Bodenlehenstraße 11 ++43-699-11108907 | A-5500 Bischofshofen | REDLINK.CO .......................................................................... | http://redlink.co/
