[jira] [Commented] (SOLR-12732) TestLogWatcher failure on Jenkins
[ https://issues.apache.org/jira/browse/SOLR-12732?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16789915#comment-16789915 ] Erick Erickson commented on SOLR-12732: --- Fixed up logging in SolrTestCaseJ4 putting a message in the first log watcher opened. I'll keep this open for another week or so to see if this is the last problem. > TestLogWatcher failure on Jenkins > - > > Key: SOLR-12732 > URL: https://issues.apache.org/jira/browse/SOLR-12732 > Project: Solr > Issue Type: Bug > Security Level: Public(Default Security Level. Issues are Public) > Components: logging >Reporter: Erick Erickson >Assignee: Erick Erickson >Priority: Minor > Attachments: SOLR-12732.patch > > > I'm 99% certain this is a test artifact, I think I see the problem. It'll > take me a lot of beasting to nail it though. > Working hypothesis is that the when we test for whether the new searcher has > no messages, we can test for no messages being logged against the watcher > before the new one _really_ gets active. -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org For additional commands, e-mail: dev-h...@lucene.apache.org
[jira] [Commented] (SOLR-12732) TestLogWatcher failure on Jenkins
[ https://issues.apache.org/jira/browse/SOLR-12732?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16789913#comment-16789913 ] ASF subversion and git services commented on SOLR-12732: Commit 6f6fbf0be2eea8373051a76bcdebf3c5019e5d6c in lucene-solr's branch refs/heads/branch_8x from Erick Erickson [ https://gitbox.apache.org/repos/asf?p=lucene-solr.git;h=6f6fbf0 ] SOLR-12732: TestLogWatcher failure on Jenkins. Deal with superclass logging bleeding into first watcher in the test > TestLogWatcher failure on Jenkins > - > > Key: SOLR-12732 > URL: https://issues.apache.org/jira/browse/SOLR-12732 > Project: Solr > Issue Type: Bug > Security Level: Public(Default Security Level. Issues are Public) > Components: logging >Reporter: Erick Erickson >Assignee: Erick Erickson >Priority: Minor > Attachments: SOLR-12732.patch > > > I'm 99% certain this is a test artifact, I think I see the problem. It'll > take me a lot of beasting to nail it though. > Working hypothesis is that the when we test for whether the new searcher has > no messages, we can test for no messages being logged against the watcher > before the new one _really_ gets active. -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org For additional commands, e-mail: dev-h...@lucene.apache.org
[jira] [Commented] (SOLR-12732) TestLogWatcher failure on Jenkins
[ https://issues.apache.org/jira/browse/SOLR-12732?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16789909#comment-16789909 ] ASF subversion and git services commented on SOLR-12732: Commit 28c8171fe6c599c2dd096503de2a60d486c1d40d in lucene-solr's branch refs/heads/master from Erick Erickson [ https://gitbox.apache.org/repos/asf?p=lucene-solr.git;h=28c8171 ] SOLR-12732: TestLogWatcher failure on Jenkins. Deal with superclass logging bleeding into first watcher in the test > TestLogWatcher failure on Jenkins > - > > Key: SOLR-12732 > URL: https://issues.apache.org/jira/browse/SOLR-12732 > Project: Solr > Issue Type: Bug > Security Level: Public(Default Security Level. Issues are Public) > Components: logging >Reporter: Erick Erickson >Assignee: Erick Erickson >Priority: Minor > Attachments: SOLR-12732.patch > > > I'm 99% certain this is a test artifact, I think I see the problem. It'll > take me a lot of beasting to nail it though. > Working hypothesis is that the when we test for whether the new searcher has > no messages, we can test for no messages being logged against the watcher > before the new one _really_ gets active. -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org For additional commands, e-mail: dev-h...@lucene.apache.org
[jira] [Commented] (SOLR-12732) TestLogWatcher failure on Jenkins
[ https://issues.apache.org/jira/browse/SOLR-12732?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16788510#comment-16788510 ] ASF subversion and git services commented on SOLR-12732: Commit 8c6e30536562413639eaf8bab1087da700733b33 in lucene-solr's branch refs/heads/master from Erick Erickson [ https://gitbox.apache.org/repos/asf?p=lucene-solr.git;h=8c6e305 ] SOLR-12732: TestLogWatcher failure on Jenkins. Added more logging > TestLogWatcher failure on Jenkins > - > > Key: SOLR-12732 > URL: https://issues.apache.org/jira/browse/SOLR-12732 > Project: Solr > Issue Type: Bug > Security Level: Public(Default Security Level. Issues are Public) > Components: logging >Reporter: Erick Erickson >Assignee: Erick Erickson >Priority: Minor > Attachments: SOLR-12732.patch > > > I'm 99% certain this is a test artifact, I think I see the problem. It'll > take me a lot of beasting to nail it though. > Working hypothesis is that the when we test for whether the new searcher has > no messages, we can test for no messages being logged against the watcher > before the new one _really_ gets active. -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org For additional commands, e-mail: dev-h...@lucene.apache.org
[jira] [Commented] (SOLR-12732) TestLogWatcher failure on Jenkins
[ https://issues.apache.org/jira/browse/SOLR-12732?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16788512#comment-16788512 ] ASF subversion and git services commented on SOLR-12732: Commit 83ab355772d46b755cab52e73051f395f938fd72 in lucene-solr's branch refs/heads/branch_8x from Erick Erickson [ https://gitbox.apache.org/repos/asf?p=lucene-solr.git;h=83ab355 ] SOLR-12732: TestLogWatcher failure on Jenkins. Added more logging (cherry picked from commit 8c6e30536562413639eaf8bab1087da700733b33) > TestLogWatcher failure on Jenkins > - > > Key: SOLR-12732 > URL: https://issues.apache.org/jira/browse/SOLR-12732 > Project: Solr > Issue Type: Bug > Security Level: Public(Default Security Level. Issues are Public) > Components: logging >Reporter: Erick Erickson >Assignee: Erick Erickson >Priority: Minor > Attachments: SOLR-12732.patch > > > I'm 99% certain this is a test artifact, I think I see the problem. It'll > take me a lot of beasting to nail it though. > Working hypothesis is that the when we test for whether the new searcher has > no messages, we can test for no messages being logged against the watcher > before the new one _really_ gets active. -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org For additional commands, e-mail: dev-h...@lucene.apache.org
[jira] [Commented] (SOLR-12732) TestLogWatcher failure on Jenkins
[ https://issues.apache.org/jira/browse/SOLR-12732?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16788217#comment-16788217 ] Erick Erickson commented on SOLR-12732: --- Oh, and I made a stupid error in the test, each loop runs for 10 seconds no matter what. I'll fix that. Don't really think it has anything to do with the test failures though. I'm beasting it and loading the machines I'm running the tests on quite heavily, but if past performance is any indication I won't be able to get it to fail anyway. If that's the case there's some additional debugging info I'll put put in. > TestLogWatcher failure on Jenkins > - > > Key: SOLR-12732 > URL: https://issues.apache.org/jira/browse/SOLR-12732 > Project: Solr > Issue Type: Bug > Security Level: Public(Default Security Level. Issues are Public) > Components: logging >Reporter: Erick Erickson >Assignee: Erick Erickson >Priority: Minor > Attachments: SOLR-12732.patch > > > I'm 99% certain this is a test artifact, I think I see the problem. It'll > take me a lot of beasting to nail it though. > Working hypothesis is that the when we test for whether the new searcher has > no messages, we can test for no messages being logged against the watcher > before the new one _really_ gets active. -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org For additional commands, e-mail: dev-h...@lucene.apache.org
[jira] [Commented] (SOLR-12732) TestLogWatcher failure on Jenkins
[ https://issues.apache.org/jira/browse/SOLR-12732?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16659013#comment-16659013 ] Erick Erickson commented on SOLR-12732: --- Rats, I still see one failure on Oct 8. > TestLogWatcher failure on Jenkins > - > > Key: SOLR-12732 > URL: https://issues.apache.org/jira/browse/SOLR-12732 > Project: Solr > Issue Type: Bug > Security Level: Public(Default Security Level. Issues are Public) > Components: logging >Reporter: Erick Erickson >Assignee: Erick Erickson >Priority: Minor > Attachments: SOLR-12732.patch > > > I'm 99% certain this is a test artifact, I think I see the problem. It'll > take me a lot of beasting to nail it though. > Working hypothesis is that the when we test for whether the new searcher has > no messages, we can test for no messages being logged against the watcher > before the new one _really_ gets active. -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org For additional commands, e-mail: dev-h...@lucene.apache.org
[jira] [Commented] (SOLR-12732) TestLogWatcher failure on Jenkins
[ https://issues.apache.org/jira/browse/SOLR-12732?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16606808#comment-16606808 ] ASF subversion and git services commented on SOLR-12732: Commit 9e04375dc193d3815e9d755514a960f902c60cd2 in lucene-solr's branch refs/heads/jira/http2 from [~cp.erick...@gmail.com] [ https://git-wip-us.apache.org/repos/asf?p=lucene-solr.git;h=9e04375 ] SOLR-12732: TestLogWatcher failure on Jenkins > TestLogWatcher failure on Jenkins > - > > Key: SOLR-12732 > URL: https://issues.apache.org/jira/browse/SOLR-12732 > Project: Solr > Issue Type: Bug > Security Level: Public(Default Security Level. Issues are Public) > Components: logging >Reporter: Erick Erickson >Assignee: Erick Erickson >Priority: Minor > Attachments: SOLR-12732.patch > > > I'm 99% certain this is a test artifact, I think I see the problem. It'll > take me a lot of beasting to nail it though. > Working hypothesis is that the when we test for whether the new searcher has > no messages, we can test for no messages being logged against the watcher > before the new one _really_ gets active. -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org For additional commands, e-mail: dev-h...@lucene.apache.org
[jira] [Commented] (SOLR-12732) TestLogWatcher failure on Jenkins
[ https://issues.apache.org/jira/browse/SOLR-12732?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16606655#comment-16606655 ] ASF subversion and git services commented on SOLR-12732: Commit dd2f85f011277a1d1170838120d9c7b7c8f34ebc in lucene-solr's branch refs/heads/branch_7x from [~cp.erick...@gmail.com] [ https://git-wip-us.apache.org/repos/asf?p=lucene-solr.git;h=dd2f85f ] SOLR-12732: TestLogWatcher failure on Jenkins (cherry picked from commit 9e04375dc193d3815e9d755514a960f902c60cd2) > TestLogWatcher failure on Jenkins > - > > Key: SOLR-12732 > URL: https://issues.apache.org/jira/browse/SOLR-12732 > Project: Solr > Issue Type: Bug > Security Level: Public(Default Security Level. Issues are Public) > Components: logging >Reporter: Erick Erickson >Assignee: Erick Erickson >Priority: Minor > Attachments: SOLR-12732.patch > > > I'm 99% certain this is a test artifact, I think I see the problem. It'll > take me a lot of beasting to nail it though. > Working hypothesis is that the when we test for whether the new searcher has > no messages, we can test for no messages being logged against the watcher > before the new one _really_ gets active. -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org For additional commands, e-mail: dev-h...@lucene.apache.org
[jira] [Commented] (SOLR-12732) TestLogWatcher failure on Jenkins
[ https://issues.apache.org/jira/browse/SOLR-12732?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16606639#comment-16606639 ] Erick Erickson commented on SOLR-12732: --- 2,000 iterations later and I still can't get this to fail (without this patch). So checking this since it's a test-only change and I'll close this JIRA after the various Jenkins jobs have had a chance to chew on it for a while. > TestLogWatcher failure on Jenkins > - > > Key: SOLR-12732 > URL: https://issues.apache.org/jira/browse/SOLR-12732 > Project: Solr > Issue Type: Bug > Security Level: Public(Default Security Level. Issues are Public) > Components: logging >Reporter: Erick Erickson >Assignee: Erick Erickson >Priority: Minor > Attachments: SOLR-12732.patch > > > I'm 99% certain this is a test artifact, I think I see the problem. It'll > take me a lot of beasting to nail it though. > Working hypothesis is that the when we test for whether the new searcher has > no messages, we can test for no messages being logged against the watcher > before the new one _really_ gets active. -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org For additional commands, e-mail: dev-h...@lucene.apache.org
[jira] [Commented] (SOLR-12732) TestLogWatcher failure on Jenkins
[ https://issues.apache.org/jira/browse/SOLR-12732?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16606640#comment-16606640 ] ASF subversion and git services commented on SOLR-12732: Commit 9e04375dc193d3815e9d755514a960f902c60cd2 in lucene-solr's branch refs/heads/master from [~cp.erick...@gmail.com] [ https://git-wip-us.apache.org/repos/asf?p=lucene-solr.git;h=9e04375 ] SOLR-12732: TestLogWatcher failure on Jenkins > TestLogWatcher failure on Jenkins > - > > Key: SOLR-12732 > URL: https://issues.apache.org/jira/browse/SOLR-12732 > Project: Solr > Issue Type: Bug > Security Level: Public(Default Security Level. Issues are Public) > Components: logging >Reporter: Erick Erickson >Assignee: Erick Erickson >Priority: Minor > Attachments: SOLR-12732.patch > > > I'm 99% certain this is a test artifact, I think I see the problem. It'll > take me a lot of beasting to nail it though. > Working hypothesis is that the when we test for whether the new searcher has > no messages, we can test for no messages being logged against the watcher > before the new one _really_ gets active. -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org For additional commands, e-mail: dev-h...@lucene.apache.org
