You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@harmony.apache.org by Mark Hindess <ma...@googlemail.com> on 2010/01/21 12:41:35 UTC
5.0M12a testing
I've created a source tar.gz for a potential 5.0M12a release by doing:
svn co https://svn.apache.org/repos/asf/harmony/enhanced/tags/5.0m12a fed
cd fed
sun60 ant populate-src bundle-src
and placed it at:
http://people.apache.org/~hindessm/5.0m12a/
Regards,
-Mark.
Re: 5.0M12a testing
Posted by Ray Chen <cl...@gmail.com>.
Hi Mark,
Thank you for answering my question, seems I misunderstood the
relationship between M12a with M13.
I didn't run the tests on windows (just for linux) when M12 released.
So maybe someone else can have a look at these test failures and
errors to see whether there is any regression.
Thank you!
--
Regards,
Ray Chen
Re: 5.0M12a testing
Posted by Mark Hindess <ma...@googlemail.com>.
In message <4d...@mail.gmail.com>, Ray
Chen writes:
>
> Hi All,
> I have run all the test of 5.0M12a (Windows XP, SP3), 25887 tests were
> run, and I see 8 failures and 7 errors as following:
[snip]
Are any of these regressions from your 5.0m12 experiences?
> My question is why JIRA6408 patch was not included ? I see the patch
> was committed @r893377 and the m12a's revision number is 901646
Because as discussed here:
http://markmail.org/thread/25csio5grqekr6cf
5.0m12a only contains a fix of a single crucial issue on top of a tag[0]
made from 5.0m12. It is not 5.0m13. 5.0m13 will be prepared for
release (from trunk) on the usual schedule.
The revision number is increased because revision numbers are a global,
monotonically increasing sequence for the entire Apache svn repository.
Regards,
Mark.
[0] mentioned in the first message in this thread:
https://svn.apache.org/repos/asf/harmony/enhanced/tags/5.0m12a
Re: 5.0M12a testing
Posted by Ray Chen <cl...@gmail.com>.
Hi All,
I have run all the test of 5.0M12a (Windows XP, SP3), 25887 tests were
run, and I see 8 failures and 7 errors as following:
Failures:
CyclicBarrierTest.testAwait2_Interrupted_BrokenBarrier
BasicScrollBarUITest.testGetThumbBounds expected: 16, actual: 0
HttpURLConnectionTest.testUsingProxy2 should throw ConnectException
(tacked by JIRA6416)
MulticastSocketTest.test_joinGroupLjava_net_InetAddress Group member
did not recv data expected:<[Hello World]> but was:<[?]>
MulticastSocketTest.test_joinGroupLjava_net_SocketAddressLjava_net_NetworkInterface
Group member did not recv data expected:<[Hello World]> but was:<[?]>
MulticastSocketTest.test_sendLjava_net_DatagramPacketB Failed to send
data. Received 512 expected:<[Hello World]> but was:<[?]>
MulticastSocketTest.test_setNetworkInterfaceLjava_net_NetworkInterface
Group member did not recv data sent on a specific interface
expected:<[{EAD768D1-33D7-4E3F-A3EE-A563356C07F0}]> but was:<[?]>
URLConnectionTest.test_getInputStream Error receiving content coded data
Errors:
CrashedTests org.apache.harmony.jpda.tests.jdwp.MultiSession.RefTypeIDTest
VM crash?
ClassObjectIDTest.testClassObjectID001 Debuggee process did not finish
during timeout
FieldAccessTest.testFieldAccess001 Debuggee process did not finish
during timeout
OutputStreamWriterTest.testHandleEarlyEOFChar_1 java.nio.BufferOverflowException
(JIRA6408)
OutputStreamWriterTest.testHandleEarlyEOFChar_2 java.nio.BufferOverflowException
(JIRA6408)
WriterTesterTest.org.apache.harmony.luni.tests.java.io.WriterTesterTest$FileWriterCharSinkTester:sinkTestWriteLargeArray java.nio.BufferOverflowException
WriterTesterTest.org.apache.harmony.luni.tests.java.io.WriterTesterTest$FileWriterCharSinkTester:sinkTestWriteLargeArray java.nio.BufferOverflowException
My question is why JIRA6408 patch was not included ? I see the patch
was committed @r893377 and the m12a's revision number is 901646
--
Regards,
Ray Chen
Re: 5.0M12a testing
Posted by Ray Chen <cl...@gmail.com>.
JarExecTest only failed when built with Visual Studio 2008
On Mon, Jan 25, 2010 at 7:43 PM, Oliver Deakin
<ol...@googlemail.com> wrote:
>
> On 23/01/2010 04:23, Nathan Beyer wrote:
>>
>> Here's my results for Windows XP SP3, Visual Studio 2008
>>
>> Two test failures. The BasicScrollBarUITest I believe is a known
>> failure; at least I've seen it before. The other failure seems
>> familiar, but I'm not sure and haven't had the time to search yet.
>>
>> BasicScrollBarUITest testGetThumbBounds Failure expected: 16,
>> actual: 0
>>
>> junit.framework.AssertionFailedError: expected: 16, actual: 0
>> at
>> javax.swing.plaf.basic.BasicScrollBarUITest.checkIsCloseTo(BasicScrollBarUITest.java:267)
>> at
>> javax.swing.plaf.basic.BasicScrollBarUITest.testGetThumbBounds(BasicScrollBarUITest.java:85)
>> at java.lang.reflect.VMReflection.invokeMethod(VMReflection.java)
>> at
>> javax.swing.BasicSwingTestCase.runBareSuper(BasicSwingTestCase.java:116)
>> at javax.swing.BasicSwingTestCase.runBareImpl(BasicSwingTestCase.java:121)
>> at javax.swing.BasicSwingTestCase$1.run(BasicSwingTestCase.java:135)
>> at java.lang.Thread.run(Thread.java:669)
>> 0.218
>>
>
> I see this failure also, but when I run BasicScrollBarUITest standalone it
> passes every time. I think we observed the same thing when we were doing M12
> testing.
>
> After a full test run on Windows XP x86 (Visual Studio 2003), apart from the
> above failure I only see 2 errors in JDWP caused by timeouts. These are
> known intermittent failures that we have covered in past releases. I don't
> see the JarExecTest failure below.
>
> Regards,
> Oliver
>
>> JarExecTest test_classpath Failure Error executing class from
>> ClassPath : result returned was incorrect.
>>
>> junit.framework.AssertionFailedError: Error executing class from
>> ClassPath : result returned was incorrect.
>> at
>> org.apache.harmony.archive.tests.java.util.jar.JarExecTest.test_classpath(JarExecTest.java:190)
>> at java.lang.reflect.VMReflection.invokeMethod(VMReflection.java)
>>
>> On Thu, Jan 21, 2010 at 5:41 AM, Mark Hindess
>> <ma...@googlemail.com> wrote:
>>
>>>
>>> I've created a source tar.gz for a potential 5.0M12a release by doing:
>>>
>>> svn co https://svn.apache.org/repos/asf/harmony/enhanced/tags/5.0m12a
>>> fed
>>> cd fed
>>> sun60 ant populate-src bundle-src
>>>
>>> and placed it at:
>>>
>>> http://people.apache.org/~hindessm/5.0m12a/
>>>
>>> Regards,
>>> -Mark.
>>>
>>>
>>>
>>>
>>
>>
>
> --
> Oliver Deakin
> 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
>
>
--
Regards,
Ray Chen
Re: 5.0M12a testing
Posted by Oliver Deakin <ol...@googlemail.com>.
On 23/01/2010 04:23, Nathan Beyer wrote:
> Here's my results for Windows XP SP3, Visual Studio 2008
>
> Two test failures. The BasicScrollBarUITest I believe is a known
> failure; at least I've seen it before. The other failure seems
> familiar, but I'm not sure and haven't had the time to search yet.
>
> BasicScrollBarUITest testGetThumbBounds Failure expected: 16, actual: 0
>
> junit.framework.AssertionFailedError: expected: 16, actual: 0
> at javax.swing.plaf.basic.BasicScrollBarUITest.checkIsCloseTo(BasicScrollBarUITest.java:267)
> at javax.swing.plaf.basic.BasicScrollBarUITest.testGetThumbBounds(BasicScrollBarUITest.java:85)
> at java.lang.reflect.VMReflection.invokeMethod(VMReflection.java)
> at javax.swing.BasicSwingTestCase.runBareSuper(BasicSwingTestCase.java:116)
> at javax.swing.BasicSwingTestCase.runBareImpl(BasicSwingTestCase.java:121)
> at javax.swing.BasicSwingTestCase$1.run(BasicSwingTestCase.java:135)
> at java.lang.Thread.run(Thread.java:669)
> 0.218
>
I see this failure also, but when I run BasicScrollBarUITest standalone
it passes every time. I think we observed the same thing when we were
doing M12 testing.
After a full test run on Windows XP x86 (Visual Studio 2003), apart from
the above failure I only see 2 errors in JDWP caused by timeouts. These
are known intermittent failures that we have covered in past releases. I
don't see the JarExecTest failure below.
Regards,
Oliver
> JarExecTest test_classpath Failure Error executing class from
> ClassPath : result returned was incorrect.
>
> junit.framework.AssertionFailedError: Error executing class from
> ClassPath : result returned was incorrect.
> at org.apache.harmony.archive.tests.java.util.jar.JarExecTest.test_classpath(JarExecTest.java:190)
> at java.lang.reflect.VMReflection.invokeMethod(VMReflection.java)
>
> On Thu, Jan 21, 2010 at 5:41 AM, Mark Hindess
> <ma...@googlemail.com> wrote:
>
>> I've created a source tar.gz for a potential 5.0M12a release by doing:
>>
>> svn co https://svn.apache.org/repos/asf/harmony/enhanced/tags/5.0m12a fed
>> cd fed
>> sun60 ant populate-src bundle-src
>>
>> and placed it at:
>>
>> http://people.apache.org/~hindessm/5.0m12a/
>>
>> Regards,
>> -Mark.
>>
>>
>>
>>
>
--
Oliver Deakin
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
Re: 5.0M12a testing
Posted by Nathan Beyer <nb...@gmail.com>.
Here's my results for Windows XP SP3, Visual Studio 2008
Two test failures. The BasicScrollBarUITest I believe is a known
failure; at least I've seen it before. The other failure seems
familiar, but I'm not sure and haven't had the time to search yet.
BasicScrollBarUITest testGetThumbBounds Failure expected: 16, actual: 0
junit.framework.AssertionFailedError: expected: 16, actual: 0
at javax.swing.plaf.basic.BasicScrollBarUITest.checkIsCloseTo(BasicScrollBarUITest.java:267)
at javax.swing.plaf.basic.BasicScrollBarUITest.testGetThumbBounds(BasicScrollBarUITest.java:85)
at java.lang.reflect.VMReflection.invokeMethod(VMReflection.java)
at javax.swing.BasicSwingTestCase.runBareSuper(BasicSwingTestCase.java:116)
at javax.swing.BasicSwingTestCase.runBareImpl(BasicSwingTestCase.java:121)
at javax.swing.BasicSwingTestCase$1.run(BasicSwingTestCase.java:135)
at java.lang.Thread.run(Thread.java:669)
0.218
JarExecTest test_classpath Failure Error executing class from
ClassPath : result returned was incorrect.
junit.framework.AssertionFailedError: Error executing class from
ClassPath : result returned was incorrect.
at org.apache.harmony.archive.tests.java.util.jar.JarExecTest.test_classpath(JarExecTest.java:190)
at java.lang.reflect.VMReflection.invokeMethod(VMReflection.java)
On Thu, Jan 21, 2010 at 5:41 AM, Mark Hindess
<ma...@googlemail.com> wrote:
>
> I've created a source tar.gz for a potential 5.0M12a release by doing:
>
> svn co https://svn.apache.org/repos/asf/harmony/enhanced/tags/5.0m12a fed
> cd fed
> sun60 ant populate-src bundle-src
>
> and placed it at:
>
> http://people.apache.org/~hindessm/5.0m12a/
>
> Regards,
> -Mark.
>
>
>
Re: 5.0M12a testing
Posted by Tim Ellison <t....@gmail.com>.
On 21/Jan/2010 13:03, Mark Hindess wrote:
> In message <4B...@gmail.com>, Tim Ellison writes:
>> On 21/Jan/2010 12:34, Mark Hindess wrote:
>>> In message <20...@d06av03.portsmouth.uk.ibm.com>,
>>> "Mark Hindess" writes:
>>>> I've created a source tar.gz for a potential 5.0M12a release by doing:
>>>>
>>>> svn co https://svn.apache.org/repos/asf/harmony/enhanced/tags/5.0m12a fed
>>>> cd fed
>>>> sun60 ant populate-src bundle-src
>>> Oops. Ignore the 'sun60' that's just a local script to set the
>>> environment to use that runtime for ant.
>> I'm building on windows at the moment.
>>
>>>> and placed it at:
>>>>
>>>> http://people.apache.org/~hindessm/5.0m12a/
>>> If you try to use this tar.gz on windows you might get errors about:
>>>
>>> working_classlib/modules/nio/src/main/java/windows
>>>
>>> and other missing directories because for some people the untar process
>>> doesn't seem to create empty directories. It is trivially fixed with
>>> mkdir but we'll need to create a source zip to vote so I'll upload one
>>> shortly.
>> I fell into that, by using WinZIP to untar the archive. If I list the
>> contents using Cygwin's GNU tar I see the empty directories are in
>> there, but they were not recreated by winzip <shrug/>.
>>
>> FYI I needed to create the following empty dirs
>> apache-harmony-5.0-src-r901653\working_classlib\modules\nio\src\main\java\windows
>> apache-harmony-5.0-src-r901653\working_classlib\modules\luni\src\test\impl\windows
>> apache-harmony-5.0-src-r901653\working_classlib\modules\suncompat\src\test\java
>
> To save anyone else this trouble I've uploaded a zip and associated files
> now.
Thanks, I'm testing these too now. So far everything looks ok.
Regards,
Tim
Re: 5.0M12a testing
Posted by Mark Hindess <ma...@googlemail.com>.
In message <4B...@gmail.com>, Tim Ellison writes:
>
> On 21/Jan/2010 12:34, Mark Hindess wrote:
> > In message <20...@d06av03.portsmouth.uk.ibm.com>,
> > "Mark Hindess" writes:
> >> I've created a source tar.gz for a potential 5.0M12a release by doing:
> >>
> >> svn co https://svn.apache.org/repos/asf/harmony/enhanced/tags/5.0m12a fed
> >> cd fed
> >> sun60 ant populate-src bundle-src
> >
> > Oops. Ignore the 'sun60' that's just a local script to set the
> > environment to use that runtime for ant.
>
> I'm building on windows at the moment.
>
> >> and placed it at:
> >>
> >> http://people.apache.org/~hindessm/5.0m12a/
> >
> > If you try to use this tar.gz on windows you might get errors about:
> >
> > working_classlib/modules/nio/src/main/java/windows
> >
> > and other missing directories because for some people the untar process
> > doesn't seem to create empty directories. It is trivially fixed with
> > mkdir but we'll need to create a source zip to vote so I'll upload one
> > shortly.
>
> I fell into that, by using WinZIP to untar the archive. If I list the
> contents using Cygwin's GNU tar I see the empty directories are in
> there, but they were not recreated by winzip <shrug/>.
>
> FYI I needed to create the following empty dirs
> apache-harmony-5.0-src-r901653\working_classlib\modules\nio\src\main\java\windows
> apache-harmony-5.0-src-r901653\working_classlib\modules\luni\src\test\impl\windows
> apache-harmony-5.0-src-r901653\working_classlib\modules\suncompat\src\test\java
To save anyone else this trouble I've uploaded a zip and associated files
now.
-Mark.
Re: 5.0M12a testing
Posted by Tim Ellison <t....@gmail.com>.
On 21/Jan/2010 12:34, Mark Hindess wrote:
> In message <20...@d06av03.portsmouth.uk.ibm.com>,
> "Mark Hindess" writes:
>> I've created a source tar.gz for a potential 5.0M12a release by doing:
>>
>> svn co https://svn.apache.org/repos/asf/harmony/enhanced/tags/5.0m12a fed
>> cd fed
>> sun60 ant populate-src bundle-src
>
> Oops. Ignore the 'sun60' that's just a local script to set the
> environment to use that runtime for ant.
I'm building on windows at the moment.
>> and placed it at:
>>
>> http://people.apache.org/~hindessm/5.0m12a/
>
> If you try to use this tar.gz on windows you might get errors about:
>
> working_classlib/modules/nio/src/main/java/windows
>
> and other missing directories because for some people the untar process
> doesn't seem to create empty directories. It is trivially fixed with
> mkdir but we'll need to create a source zip to vote so I'll upload one
> shortly.
I fell into that, by using WinZIP to untar the archive. If I list the
contents using Cygwin's GNU tar I see the empty directories are in
there, but they were not recreated by winzip <shrug/>.
FYI I needed to create the following empty dirs
apache-harmony-5.0-src-r901653\working_classlib\modules\nio\src\main\java\windows
apache-harmony-5.0-src-r901653\working_classlib\modules\luni\src\test\impl\windows
apache-harmony-5.0-src-r901653\working_classlib\modules\suncompat\src\test\java
Regards,
Tim
Re: 5.0M12a testing
Posted by Mark Hindess <ma...@googlemail.com>.
In message <20...@d06av03.portsmouth.uk.ibm.com>,
"Mark Hindess" writes:
>
> I've created a source tar.gz for a potential 5.0M12a release by doing:
>
> svn co https://svn.apache.org/repos/asf/harmony/enhanced/tags/5.0m12a fed
> cd fed
> sun60 ant populate-src bundle-src
Oops. Ignore the 'sun60' that's just a local script to set the
environment to use that runtime for ant.
> and placed it at:
>
> http://people.apache.org/~hindessm/5.0m12a/
If you try to use this tar.gz on windows you might get errors about:
working_classlib/modules/nio/src/main/java/windows
and other missing directories because for some people the untar process
doesn't seem to create empty directories. It is trivially fixed with
mkdir but we'll need to create a source zip to vote so I'll upload one
shortly.
Regards,
-Mark