[jira] [Commented] (SOLR-3461) TestRealTimeGet.testStressRecovery() is sometimes very slow on Windows, using al CPUs
[ https://issues.apache.org/jira/browse/SOLR-3461?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13279678#comment-13279678 ] Dawid Weiss commented on SOLR-3461: --- Thanks Uwe. I'll try to dig on a vm in a spare minute to rule out any test framework interference (although I doubt it's the cause). TestRealTimeGet.testStressRecovery() is sometimes very slow on Windows, using al CPUs - Key: SOLR-3461 URL: https://issues.apache.org/jira/browse/SOLR-3461 Project: Solr Issue Type: Bug Components: update Affects Versions: 4.0 Reporter: Uwe Schindler I already noticed that on my local machine (Thinkpad Laptop with SSD), but I was thinking it might be my slow IO system (thanks Robert for this running-gag). But when reviewing test times on the new Jenkins Windows build server, this test takes very long (around 30 minutes, which is half of the complete time when running with 1 test runner on 2 cores). There must be something that makes this test very slow on Windows: http://goo.gl/irDVw The method taking so long is actually testStressRecovery, I would like to @Ignore it until it is fixed. It makes running Solr tests in Windows take ages and stops me from running them at all :( -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira - To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org For additional commands, e-mail: dev-h...@lucene.apache.org
[jira] [Commented] (SOLR-3461) TestRealTimeGet.testStressRecovery() is sometimes very slow on Windows, using al CPUs
[ https://issues.apache.org/jira/browse/SOLR-3461?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13279472#comment-13279472 ] Dawid Weiss commented on SOLR-3461: --- I ran the above 20 times (windows 7, 64 bit, 4 ht cores (8 logical)). All times were sensible: {noformat} Total time: 3 minutes 50 seconds Total time: 3 minutes 43 seconds Total time: 3 minutes 49 seconds Total time: 3 minutes 45 seconds Total time: 3 minutes 27 seconds Total time: 3 minutes 50 seconds Total time: 4 minutes 9 seconds Total time: 3 minutes 42 seconds Total time: 3 minutes 35 seconds Total time: 3 minutes 55 seconds Total time: 3 minutes 36 seconds Total time: 3 minutes 42 seconds Total time: 4 minutes 8 seconds Total time: 3 minutes 54 seconds Total time: 3 minutes 15 seconds Total time: 4 minutes 2 seconds Total time: 3 minutes 48 seconds Total time: 3 minutes 57 seconds Total time: 3 minutes 49 seconds Total time: 3 minutes 44 seconds {noformat} I did encounter two failures though: {noformat} [junit4] Suite: org.apache.solr.cloud.OverseerTest [junit4] (@BeforeClass output) [junit4] 2 Creating dataDir: c:\Work\lucene-solr\solr\build\solr-core\test\J2\.\solrtest-OverseerTest-1337375903710 [junit4] 2 [junit4] FAILURE 16.5s J2 | OverseerTest.testShardLeaderChange [junit4] Throwable #1: org.junit.ComparisonFailure: Unexpected shard leader coll:collection1 shard:shard1 expected:core[1] but was:core[4] [junit4]at __randomizedtesting.SeedInfo.seed([4877EC7806E56AF3:96246B8F1C7D9F02]:0) [junit4]at org.junit.Assert.assertEquals(Assert.java:125) [junit4]at org.apache.solr.cloud.OverseerTest.verifyShardLeader(OverseerTest.java:550) [junit4]at org.apache.solr.cloud.OverseerTest.testShardLeaderChange(OverseerTest.java:709) [junit4]at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) [junit4]at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57) [junit4]at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) [junit4]at java.lang.reflect.Method.invoke(Method.java:601) [junit4]at com.carrotsearch.randomizedtesting.RandomizedRunner.invoke(RandomizedRunner.java:1969) [junit4]at com.carrotsearch.randomizedtesting.RandomizedRunner.access$1100(RandomizedRunner.java:132) [junit4]at com.carrotsearch.randomizedtesting.RandomizedRunner$6.evaluate(RandomizedRunner.java:814) [junit4]at com.carrotsearch.randomizedtesting.RandomizedRunner$7.evaluate(RandomizedRunner.java:875) [junit4]at com.carrotsearch.randomizedtesting.RandomizedRunner$8.evaluate(RandomizedRunner.java:889) [junit4]at com.carrotsearch.randomizedtesting.rules.SystemPropertiesRestoreRule$1.evaluate(SystemPropertiesRestoreRule.java:53) [junit4]at org.apache.lucene.util.TestRuleSetupTeardownChained$1.evaluate(TestRuleSetupTeardownChained.java:50) [junit4]at org.apache.lucene.util.TestRuleFieldCacheSanity$1.evaluate(TestRuleFieldCacheSanity.java:32) [junit4]at org.apache.lucene.util.AbstractBeforeAfterRule$1.evaluate(AbstractBeforeAfterRule.java:45) [junit4]at com.carrotsearch.randomizedtesting.rules.SystemPropertiesInvariantRule$1.evaluate(SystemPropertiesInvariantRule.java:55) [junit4]at org.apache.lucene.util.TestRuleReportUncaughtExceptions$1.evaluate(TestRuleReportUncaughtExceptions.java:68) [junit4]at org.apache.lucene.util.TestRuleThreadAndTestName$1.evaluate(TestRuleThreadAndTestName.java:48) [junit4]at org.apache.lucene.util.TestRuleMarkFailure$1.evaluate(TestRuleMarkFailure.java:48) [junit4]at com.carrotsearch.randomizedtesting.RandomizedRunner.runSingleTest(RandomizedRunner.java:821) [junit4]at com.carrotsearch.randomizedtesting.RandomizedRunner.access$700(RandomizedRunner.java:132) [junit4]at com.carrotsearch.randomizedtesting.RandomizedRunner$3$1.run(RandomizedRunner.java:669) [junit4]at com.carrotsearch.randomizedtesting.RandomizedRunner$3.evaluate(RandomizedRunner.java:695) [junit4]at com.carrotsearch.randomizedtesting.RandomizedRunner$4.evaluate(RandomizedRunner.java:734) [junit4]at com.carrotsearch.randomizedtesting.RandomizedRunner$5.evaluate(RandomizedRunner.java:745) [junit4]at com.carrotsearch.randomizedtesting.rules.SystemPropertiesRestoreRule$1.evaluate(SystemPropertiesRestoreRule.java:53) [junit4]at org.apache.lucene.util.AbstractBeforeAfterRule$1.evaluate(AbstractBeforeAfterRule.java:45) [junit4]at org.apache.lucene.util.TestRuleReportUncaughtExceptions$1.evaluate(TestRuleReportUncaughtExceptions.java:68)
[jira] [Commented] (SOLR-3461) TestRealTimeGet.testStressRecovery() is sometimes very slow on Windows, using al CPUs
[ https://issues.apache.org/jira/browse/SOLR-3461?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13279514#comment-13279514 ] Uwe Schindler commented on SOLR-3461: - Hi Dawid, it looks like this test is slow on Windows (maybe also Linux) systems with 2 CPU cores. This is why I said some starvation or whatever. For me it reproduces to be slow in lots of cases on 3 different Windows 7 machines (2 Laptops and one VirtualBOX), each with 2 CPUs. I can almost always reproduce it here locally. I dont understand the whole test so that is all I can say! TestRealTimeGet.testStressRecovery() is sometimes very slow on Windows, using al CPUs - Key: SOLR-3461 URL: https://issues.apache.org/jira/browse/SOLR-3461 Project: Solr Issue Type: Bug Components: update Affects Versions: 4.0 Reporter: Uwe Schindler I already noticed that on my local machine (Thinkpad Laptop with SSD), but I was thinking it might be my slow IO system (thanks Robert for this running-gag). But when reviewing test times on the new Jenkins Windows build server, this test takes very long (around 30 minutes, which is half of the complete time when running with 1 test runner on 2 cores). There must be something that makes this test very slow on Windows: http://goo.gl/irDVw The method taking so long is actually testStressRecovery, I would like to @Ignore it until it is fixed. It makes running Solr tests in Windows take ages and stops me from running them at all :( -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira - To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org For additional commands, e-mail: dev-h...@lucene.apache.org
[jira] [Commented] (SOLR-3461) TestRealTimeGet.testStressRecovery() is sometimes very slow on Windows, using al CPUs
[ https://issues.apache.org/jira/browse/SOLR-3461?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13279515#comment-13279515 ] Uwe Schindler commented on SOLR-3461: - The sugester bugs also happened on the new Jenkins VM. TestRealTimeGet.testStressRecovery() is sometimes very slow on Windows, using al CPUs - Key: SOLR-3461 URL: https://issues.apache.org/jira/browse/SOLR-3461 Project: Solr Issue Type: Bug Components: update Affects Versions: 4.0 Reporter: Uwe Schindler I already noticed that on my local machine (Thinkpad Laptop with SSD), but I was thinking it might be my slow IO system (thanks Robert for this running-gag). But when reviewing test times on the new Jenkins Windows build server, this test takes very long (around 30 minutes, which is half of the complete time when running with 1 test runner on 2 cores). There must be something that makes this test very slow on Windows: http://goo.gl/irDVw The method taking so long is actually testStressRecovery, I would like to @Ignore it until it is fixed. It makes running Solr tests in Windows take ages and stops me from running them at all :( -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira - To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org For additional commands, e-mail: dev-h...@lucene.apache.org
[jira] [Commented] (SOLR-3461) TestRealTimeGet.testStressRecovery() is sometimes very slow on Windows, using al CPUs
[ https://issues.apache.org/jira/browse/SOLR-3461?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13277687#comment-13277687 ] Uwe Schindler commented on SOLR-3461: - It is not always slow, but I have seen the 30 minute runs now on 3 different PCs running: - 64 bit Java 1.6.0_32 or 1.7.0_u4 Server JVM - Windows 7 Professional - = 2 CPUs So this needs fixing, sorry. Maybe you don't see the issue there is still something wrong. TestRealTimeGet.testStressRecovery() is sometimes very slow on Windows, using al CPUs - Key: SOLR-3461 URL: https://issues.apache.org/jira/browse/SOLR-3461 Project: Solr Issue Type: Bug Components: update Affects Versions: 4.0 Reporter: Uwe Schindler I already noticed that on my local machine (Thinkpad Laptop with SSD), but I was thinking it might be my slow IO system (thanks Robert for this running-gag). But when reviewing test times on the new Jenkins Windows build server, this test takes very long (around 30 minutes, which is half of the complete time when running with 1 test runner on 2 cores). There must be something that makes this test very slow on Windows: http://goo.gl/irDVw The method taking so long is actually testStressRecovery, I would like to @Ignore it until it is fixed. It makes running Solr tests in Windows take ages and stops me from running them at all :( -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira - To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org For additional commands, e-mail: dev-h...@lucene.apache.org
[jira] [Commented] (SOLR-3461) TestRealTimeGet.testStressRecovery() is sometimes very slow on Windows, using al CPUs
[ https://issues.apache.org/jira/browse/SOLR-3461?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13277688#comment-13277688 ] Uwe Schindler commented on SOLR-3461: - It is slow in 80% of all cases for the following test seed: ant test -Dtests.seed=4877EC7806E56AF3 TestRealTimeGet.testStressRecovery() is sometimes very slow on Windows, using al CPUs - Key: SOLR-3461 URL: https://issues.apache.org/jira/browse/SOLR-3461 Project: Solr Issue Type: Bug Components: update Affects Versions: 4.0 Reporter: Uwe Schindler I already noticed that on my local machine (Thinkpad Laptop with SSD), but I was thinking it might be my slow IO system (thanks Robert for this running-gag). But when reviewing test times on the new Jenkins Windows build server, this test takes very long (around 30 minutes, which is half of the complete time when running with 1 test runner on 2 cores). There must be something that makes this test very slow on Windows: http://goo.gl/irDVw The method taking so long is actually testStressRecovery, I would like to @Ignore it until it is fixed. It makes running Solr tests in Windows take ages and stops me from running them at all :( -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira - To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org For additional commands, e-mail: dev-h...@lucene.apache.org
[jira] [Commented] (SOLR-3461) TestRealTimeGet.testStressRecovery() is sometimes very slow on Windows, using al CPUs
[ https://issues.apache.org/jira/browse/SOLR-3461?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13276777#comment-13276777 ] Uwe Schindler commented on SOLR-3461: - I investigated: It is not related to Codecs or something like that, testStressRecovery uses Solr's API and not Lucene's API. It sometimes is very slow, this cannot always reproduced with the seed. So there must be some Windows-only deadlock/starvation-like problem, or maybe related to 2 cores like on the Jenkins Windows machine or my local laptop. I added an assumeFalse(Constants.WINDOWS) for now. TestRealTimeGet.testStressRecovery() is sometimes very slow on Windows, using al CPUs - Key: SOLR-3461 URL: https://issues.apache.org/jira/browse/SOLR-3461 Project: Solr Issue Type: Bug Components: update Affects Versions: 4.0 Reporter: Uwe Schindler I already noticed that on my local machine (Thinkpad Laptop with SSD), but I was thinking it might be my slow IO system (thanks Robert for this running-gag). But when reviewing test times on the new Jenkins Windows build server, this test takes very long (around 30 minutes, which is half of the complete time when running with 1 test runner on 2 cores). There must be something that makes this test very slow on Windows: http://goo.gl/irDVw The method taking so long is actually testStressRecovery, I would like to @Ignore it until it is fixed. It makes running Solr tests in Windows take ages and stops me from running them at all :( -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira - To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org For additional commands, e-mail: dev-h...@lucene.apache.org
[jira] [Commented] (SOLR-3461) TestRealTimeGet.testStressRecovery() is sometimes very slow on Windows, using al CPUs
[ https://issues.apache.org/jira/browse/SOLR-3461?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13276779#comment-13276779 ] Uwe Schindler commented on SOLR-3461: - Committed test supression on windows at revision: 1339209 TestRealTimeGet.testStressRecovery() is sometimes very slow on Windows, using al CPUs - Key: SOLR-3461 URL: https://issues.apache.org/jira/browse/SOLR-3461 Project: Solr Issue Type: Bug Components: update Affects Versions: 4.0 Reporter: Uwe Schindler I already noticed that on my local machine (Thinkpad Laptop with SSD), but I was thinking it might be my slow IO system (thanks Robert for this running-gag). But when reviewing test times on the new Jenkins Windows build server, this test takes very long (around 30 minutes, which is half of the complete time when running with 1 test runner on 2 cores). There must be something that makes this test very slow on Windows: http://goo.gl/irDVw The method taking so long is actually testStressRecovery, I would like to @Ignore it until it is fixed. It makes running Solr tests in Windows take ages and stops me from running them at all :( -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira - To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org For additional commands, e-mail: dev-h...@lucene.apache.org
[jira] [Commented] (SOLR-3461) TestRealTimeGet.testStressRecovery() is sometimes very slow on Windows, using al CPUs
[ https://issues.apache.org/jira/browse/SOLR-3461?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13277230#comment-13277230 ] Jack Krupansky commented on SOLR-3461: -- It runs reasonably on my Windows 7 i5-based notebook PC: [junit4] Suite: org.apache.solr.search.TestRealTimeGet [junit4] Completed on J1 in 130.37s, 8 tests TestRealTimeGet.testStressRecovery() is sometimes very slow on Windows, using al CPUs - Key: SOLR-3461 URL: https://issues.apache.org/jira/browse/SOLR-3461 Project: Solr Issue Type: Bug Components: update Affects Versions: 4.0 Reporter: Uwe Schindler I already noticed that on my local machine (Thinkpad Laptop with SSD), but I was thinking it might be my slow IO system (thanks Robert for this running-gag). But when reviewing test times on the new Jenkins Windows build server, this test takes very long (around 30 minutes, which is half of the complete time when running with 1 test runner on 2 cores). There must be something that makes this test very slow on Windows: http://goo.gl/irDVw The method taking so long is actually testStressRecovery, I would like to @Ignore it until it is fixed. It makes running Solr tests in Windows take ages and stops me from running them at all :( -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira - To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org For additional commands, e-mail: dev-h...@lucene.apache.org
[jira] [Commented] (SOLR-3461) TestRealTimeGet.testStressRecovery() is sometimes very slow on Windows, using al CPUs
[ https://issues.apache.org/jira/browse/SOLR-3461?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13277269#comment-13277269 ] Jack Krupansky commented on SOLR-3461: -- With the change, the test went from 130 seconds to 20. Even though 130 sec is nowhere near as slow as your 30 minutes, 2 minutes is still a long time. It was taking 10 seconds in the nightly build yesterday. From my latest local test run: [junit4] Suite: org.apache.solr.search.TestRealTimeGet [junit4] IGNOR/A 0.04s J1 | TestRealTimeGet.testStressRecovery [junit4] Assumption #1: FIXME: This test is horribly slow sometimes on Windows! [junit4] Completed on J1 in 19.24s, 8 tests, 1 skipped TestRealTimeGet.testStressRecovery() is sometimes very slow on Windows, using al CPUs - Key: SOLR-3461 URL: https://issues.apache.org/jira/browse/SOLR-3461 Project: Solr Issue Type: Bug Components: update Affects Versions: 4.0 Reporter: Uwe Schindler I already noticed that on my local machine (Thinkpad Laptop with SSD), but I was thinking it might be my slow IO system (thanks Robert for this running-gag). But when reviewing test times on the new Jenkins Windows build server, this test takes very long (around 30 minutes, which is half of the complete time when running with 1 test runner on 2 cores). There must be something that makes this test very slow on Windows: http://goo.gl/irDVw The method taking so long is actually testStressRecovery, I would like to @Ignore it until it is fixed. It makes running Solr tests in Windows take ages and stops me from running them at all :( -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira - To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org For additional commands, e-mail: dev-h...@lucene.apache.org