I suspect that the PermissionCollectionTest failure began with this commit - http://svn.apache.org/viewvc?view=rev&revision=612718. As of this commit, SimpleTimeZone began delegating to ICU's TimeZone classes, which end up loading the class ICUDebug [1], which performs a System.getProperty("java.version") inline without catching a SecurityException.
It looks like we need to add a privileged block around this bit of code. [1] http://source.icu-project.org/repos/icu/icu4j/tags/release-3-8-1/src/com/ibm/icu/impl/ICUDebug.java On Mon, Aug 18, 2008 at 9:04 PM, chunrong lai <[EMAIL PROTECTED]> wrote: > I have not checked when the PermissionCollectionTest issue started but I can > see the failure in M6 so I think it is not a new bug. > http://people.apache.org/~smishura/r653525/Linux_x86_64/classlib-test/tests/api/java/security/1390_PermissionCollectionTest.html<http://people.apache.org/~smishura/r653525/Linux_x86_64/classlib-test/> > > The page http://people.apache.org/~chunrong/harmony-integrity/ includes the > integrity testing results of the four platforms, with less test suites than > the snapshot testing cycle. I see that the results of win32 is not worse > than the other three platforms there. > > > On 8/19/08, Nathan Beyer <[EMAIL PROTECTED]> wrote: >> >> Is there a bug logged for the PermissionCollectionTest issue? >> >> Additionally, I thought you mentioned that you had a linux x86_64 box >> running. The integration test page that's referenced for this vote >> shows Windows 64 and Linux 32. What about Windows 32? >> >> -Nathan >> >> On Sun, Aug 17, 2008 at 11:51 PM, chunrong lai <[EMAIL PROTECTED]> >> wrote: >> > Thanks for the sharing. It is strange that I did not observe the >> unpack200 >> > failures. >> > Other classlib failures are not new. The guity commit of the management >> > failures has been identified as r644719. >> > The issues in jdktools-test are "timeouts" in our testing. They are not >> new >> > either. Moreover the standalone running, for example "java -cp >> > >> "/home/laichunrong/apache-ant-1.6.5/lib/junit.jar:./infra/build/checkouts/jdktools/build/tests/classes/" >> > org.apache.harmony.jpda.tests.jdwp.MultiSession.ResumeTest" just report >> "OK" >> > here (should set LD_LIBRARY_PATH to include >> > >> ./infra/build/checkouts/jdktools/modules/jpda/src/main/native/jdwp/unix/transport). >> > So >> > I do not think that they are blockers for now. >> > Thanks. >> > >> > On 8/18/08, Nathan Beyer <[EMAIL PROTECTED]> wrote: >> >> >> >> -1 I'm not able to reproduce the same results on x86_64 Ubuntu Hardy >> >> >> >> Here are the failures I'm seeing, consistently, in classlib-test. The >> >> unpack200 test failures concern me. >> >> >> >> error testGetAttribute >> >> org.apache.harmony.lang.management.MemoryPoolMXBeanImplTest >> >> error testGetCollectionUsage >> >> >> >> >> org.apache.harmony.lang.management.tests.java.lang.management.MemoryPoolMXBeanTest >> >> failure testWithSql >> >> org.apache.harmony.unpack200.tests.ArchiveTest >> >> failure >> >> testHelloWorld org.apache.harmony.unpack200.tests.SegmentTest >> >> failure test_impliesLjava_security_Permission >> >> tests.api.java.security.PermissionCollectionTest >> >> >> >> Here are the failures I'm seeing, consistently, in jdktools-test. >> >> >> >> error testDebuggerLaunch001 >> >> >> >> >> org.apache.harmony.jpda.tests.jdwp.DebuggerOnDemand.OnthrowDebuggerLaunchTest >> >> error testDebuggerLaunch002 >> >> >> >> >> org.apache.harmony.jpda.tests.jdwp.DebuggerOnDemand.OnthrowDebuggerLaunchTest >> >> error testDebuggerLaunch003 >> >> >> >> >> org.apache.harmony.jpda.tests.jdwp.DebuggerOnDemand.OnthrowDebuggerLaunchTest >> >> error testDebuggerLaunch004 >> >> >> >> >> org.apache.harmony.jpda.tests.jdwp.DebuggerOnDemand.OnthrowDebuggerLaunchTest >> >> error testAttachConnector001 >> >> >> org.apache.harmony.jpda.tests.jdwp.MultiSession.AttachConnectorTest >> >> error testClearBreakpoint001 >> >> org.apache.harmony.jpda.tests.jdwp.MultiSession.BreakpointTest >> >> error testMethodEvent001 >> >> >> org.apache.harmony.jpda.tests.jdwp.MultiSession.MethodEntryExitTest >> >> error >> >> >> testResume org.apache.harmony.jpda.tests.jdwp.MultiSession.ResumeTest >> >> error testThreadEnd001 >> >> org.apache.harmony.jpda.tests.jdwp.MultiSession.ThreadEndTest >> >> error testThreadStart001 >> >> org.apache.harmony.jpda.tests.jdwp.MultiSession.ThreadStartTest >> >> >> >> drlvm-test and drlvm-reg-test are passing. >> >> >> >> -Nathan >> >> >> >> On Fri, Aug 15, 2008 at 8:26 AM, Sian January >> >> <[EMAIL PROTECTED]> wrote: >> >> > Hi everyone, >> >> > >> >> > We have completed a testing cycle for r681495 and evaluated the >> results >> >> [1]. >> >> > >> >> > There are some test failures, but most of them are either intermittent >> >> > or were also failing in M6. However if anyone thinks any of these >> >> > test failures are blockers and has not yet had a chance to say, please >> >> > speak up now. >> >> > >> >> > Otherwise, shall we declare r681495 as M7? >> >> > >> >> > Thanks, >> >> > >> >> > Sian >> >> > >> >> > >> >> > [1] http://harmony.markmail.org/message/cpfcnslv53doueeg?q= >> >> > >> >> > >> >> > -- >> >> > Unless stated otherwise above: >> >> > IBM United Kingdom Limited - Registered in England and Wales with >> number >> >> 741598. >> >> > Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 >> >> 3AU >> >> > >> >> >> > >> >
