Bug 1100839
| Summary: | (6.4.0) Intermittent fail of some datasource tests on IBM JDK after attempt to load mysql driver for h2 datasource | ||||||
|---|---|---|---|---|---|---|---|
| Product: | [JBoss] JBoss Enterprise Application Platform 6 | Reporter: | Pavel Jelinek <pjelinek> | ||||
| Component: | Testsuite | Assignee: | Panagiotis Sotiropoulos <psotirop> | ||||
| Status: | CLOSED WONTFIX | QA Contact: | Petr Kremensky <pkremens> | ||||
| Severity: | medium | Docs Contact: | |||||
| Priority: | unspecified | ||||||
| Version: | 6.3.0 | CC: | bbaranow, cdewolf, kkhan, pkremens, psotirop, thofman | ||||
| Target Milestone: | --- | ||||||
| Target Release: | --- | ||||||
| Hardware: | Unspecified | ||||||
| OS: | Unspecified | ||||||
| Whiteboard: | |||||||
| Fixed In Version: | Doc Type: | Bug Fix | |||||
| Doc Text: | Story Points: | --- | |||||
| Clone Of: | |||||||
| : | 1135963 (view as bug list) | Environment: |
IBMJDK
|
||||
| Last Closed: | 2015-03-23 13:47:02 UTC | Type: | Bug | ||||
| Regression: | --- | Mount Type: | --- | ||||
| Documentation: | --- | CRM: | |||||
| Verified Versions: | Category: | --- | |||||
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |||||
| Cloudforms Team: | --- | Target Upstream Version: | |||||
| Embargoed: | |||||||
| Bug Depends On: | |||||||
| Bug Blocks: | 996500, 1135963 | ||||||
| Attachments: |
|
||||||
|
Description
Pavel Jelinek
2014-05-23 14:49:01 UTC
Seen also with 6.3.1.CP.CR1 https://jenkins.mw.lab.eng.bos.redhat.com/hudson/job/eap-63x-patched-testsuite-rhel/7/jdk=ibm1.6,label_exp=eap-sustaining%20&&%20RHEL5%20&&%20x86/ Tested with IBM JDK 1.6 and the tests pass. Could the QA rerun the tests and provide an updated status? Hello, yes, we can still see it, but it doesn't occurs every time and I was not able to find what's triggering it so far. https://jenkins.mw.lab.eng.bos.redhat.com/hudson/job/eap-6x-as-testsuite-rhel/29/RELEASE=6.4.0,jdk=ibm1.6,label_exp=eap-sustaining&&RHEL5&&x86/testReport/junit/ Try this link instead please: https://jenkins.mw.lab.eng.bos.redhat.com/hudson/job/eap-6x-as-testsuite-rhel/29/RELEASE=6.4.0,jdk=ibm1.6,label_exp=eap-sustaining%20&&%20RHEL5%20&&%20x86/testReport/ I have run the testsuite for several times and all the tests pass. Which build of the IBM JDK 1.6 is used? The testsuite was run 10 times (all tests pass) : using Fedora release 20 (Heisenbug) and java version "1.6.0" [ IBM J9 VM (build 2.4, JRE 1.6.0 IBM J9 2.4 Linux amd64-64 jvmxa6460sr16-20140626_204542 (JIT enabled, AOT enabled) J9VM - 20140626_204542 JIT - r9_20130920_46510ifx7 GC - GA24_Java6_SR16_20140626_1848_B204542) ] Could I have a clone of this job to run it on Jenkins? The problem with all these intermittent failures is that they occurs only from time to time so it's a quite challenge to find a reproducer sometimes (and this one I see quite rarely). We are using following IBM build: 1.6.0 build 20140418_01 (SR16) I've created a new job you can use (feel free to make any configuration changes). https://jenkins.mw.lab.eng.bos.redhat.com/hudson/job/eap-640-as-testsuite-bz1100839-datasource-smoke-ibm/ however Jenkins is a little bit overloaded so no results yet. I'll try to run it few times to see we can run into it. FYI: *prepare.zip used by job contains compiled sources, local maven repo and jboss distribution. Could I have permission to upload another IBM JDK 1.6 build on Jenkins? Do you notice the same issue also when you build wildfly? You can upload jdk by yourself (use /home/hudson/users-tmp/$USER_NAME folder) and set it on PATH manually in job's configuration (this way you don't have to create any tickets and wait). We don't run any CI jobs with WF. You may ask people around WF like Stuart Douglas or so. Panagiotis Sotiropoulos <panossot> updated the status of jira JBTEST-21 to Closed Adding <runOrder>alphabetical</runOrder> in the surefire configuration in the smoke pom seams to be a workaround. thx Petr for the help Caused by: java.lang.NoClassDefFoundError: com/mysql/jdbc/StringUtils at com.mysql.jdbc.NonRegisteringDriver.parseURL(NonRegisteringDriver.java:582) at com.mysql.jdbc.NonRegisteringDriver.acceptsURL(NonRegisteringDriver.java:236) at java.sql.DriverManager.getDriver(DriverManager.java:495) [sql.jar:] at org.jboss.jca.adapters.jdbc.local.LocalManagedConnectionFactory.isDriverLoadedForURL(LocalManagedConnectionFactory.java:671) at org.jboss.jca.adapters.jdbc.local.LocalManagedConnectionFactory.getDriver(LocalManagedConnectionFactory.java:645) at org.jboss.jca.adapters.jdbc.local.LocalManagedConnectionFactory.getLocalManagedConnection(LocalManagedConnectionFactory.java:320) DriverManager.getDriver(url) in LocalManagedConnectionFactory:671 is given url argument "jdbc:h2:mem:test;DB_CLOSE_DELAY=-1;DB_CLOSE_ON_EXIT=FALSE", while it produces exception about com.mysql.jdbc.StringUtils. That seems like that is a bug in IBM JDK. PR of workaround sent : https://github.com/jbossas/jboss-eap/pull/2044 Panagiotis Sotiropoulos <panossot> updated the status of jira JBQA-10003 to Coding In Progress Panagiotis Sotiropoulos <panossot> updated the status of jira JBQA-10003 to Open Panagiotis Sotiropoulos <panossot> updated the status of jira JBQA-10003 to Resolved The error is caused inside the IBM JDK 1.6 DriverManager.getDriver class. An exception is not caught by isDriverLoadedForURL class (ironjacamar - adapters). Code : driver = (Driver)clazz.newInstance(); DriverManager.registerDriver(driver); is never executed. If tested with unmanaged container all the tests succeed. PR sent : https://github.com/ironjacamar/ironjacamar/pull/304 Fixes the IBM JDK 1.6 issue by getting the LocalManagedConnection after reloading the driver when the first try fails. Triage: it is not a common occurrence to install a new driver version and not expect a server restart (i.c.w. IBM JDK 6) |