db-derby-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Myrna van Lunteren <m.v.lunte...@gmail.com>
Subject Re: [jira] Commented: (DERBY-4785) Remove JCC tests and references to JCC in test code
Date Wed, 15 Dec 2010 18:29:41 GMT
On Tue, Dec 14, 2010 at 7:36 PM, Jayaram Subramanian
<rsjay1976@gmail.com> wrote:
> Hi All,
> I see the following in the T_RawStoreFactory.diff file.. Please let me
> know any corrective actions if required in my setup..
> *** Start: T_RawStoreFactory jdk1.6.0 storeall:storeunit 2010-12-06 23:08:36 ***
> 2 del
> < -- Unit Test T_RawStoreFactory finished
> 2 add
>> Shutting down due to unit test failure.
> Test Failed.
> *** End:   T_RawStoreFactory jdk1.6.0 storeall:storeunit 2010-12-06 23:09:12 ***
> With Regards
> Jayaram
> On Sat, Dec 11, 2010 at 10:24 AM, Bryan Pendleton
> <bpendleton.derby@gmail.com> wrote:
>> On 12/11/2010 08:02 AM, Jayaram Subramanian wrote:
>>> Thanks for your reply.. When i run DerbyAll from a clean folder I dont
>>> see any errors in derbyall.sum file..
>>> But in derbyall.fail file i see the following
>>> ----------------------------------------------------------
>>> derbyall/storeall/storeall.fail:unit/T_RawStoreFactory.unit
>> This test seems to intermittently fail, in a variety of environments, for
>> example see https://issues.apache.org/jira/browse/DERBY-3993
>> It is possible that the test failure you are seeing is due to
>> this intermittent failure.
>> In your folder where you ran the tests, are there any other
>> files which contain more information about the T_RawStoreFactory failure?
>> In particular, look for a 'storeunit' directory, and look for a file
>> such as storeunit/T_RawStoreFactory.diff
>> If you have a file like that, what are the contents?
>> thanks,
>> bryan


As Bryan said, this test fails intermittently in a variety of
environments. There's probably some timing issue somewhere in the
tests (or in the Derby source code). As you can tell from the
'probably', we don't know exactly how to get around this; so, you need
to figure out if this failure is due to a change you made, or similar
to the other failures reported.

There are a couple of things you can do to figure out if you need to
worry about this (or any) failure:
- run the test again, by itself (this particular old harness can be
run so: java org.apache.derbyTesting.functionTests.harness.RunTest
unit/T_RawStoreFactory.unit) ; or make a script to run it 20 times.
  Unfortunately, with an intermittent and likely timing-related issue,
this result is not really anything conclusive, merely an info-point.
- look at all other test files. These 'old harness' tests leave all
relevant files behind if the test fails. So, you should be able to
find a .tmp (actual output) .out (modified output), .tmpmstr (expected
output) files, and a derby.log file.
    If comparing tmpmstr and .tmp, nor the derby.log file seem related
to your changes, than it's likely that you're running into something
similar to DERBY-3993.


View raw message