probably not related to the repackaging. I have a candidate here [0], it
looks like this file is not closed, which may account for Win complaining.
let me see if I can fix it.



[0]
https://github.com/apache/jackrabbit-oak/blob/trunk/oak-segment/src/test/java/org/apache/jackrabbit/oak/plugins/segment/file/FileStoreIT.java#L188

On Fri, Jan 22, 2016 at 3:03 PM, Julian Reschke <
[email protected]> wrote:

> On 2016-01-22 14:57, Francesco Mari wrote:
>
>> 2016-01-22 14:22 GMT+01:00 Julian Reschke <[email protected]>:
>>
>> On 2016-01-21 19:30, Francesco Mari wrote:
>>>
>>> Hi all,
>>>>
>>>> I committed a big change in r1726048 because of OAK-3744. I had to move
>>>> some code around and adjust some POMs and tests. Please let me know if
>>>> something is broken for you. I did my best by running unit and
>>>> integration
>>>> tests on my machine before committing, but I can't guarantee a 100%
>>>> success
>>>> rate on every development machine out there.
>>>>
>>>> If something is broken, please reply to this email or create a new issue
>>>> and assign it to me.
>>>>
>>>> Thanks,
>>>>
>>>> Francesco
>>>>
>>>>
>>> Hi there,
>>>
>>> I see two problems.
>>>
>>> 1) integration tests failing:
>>>
>>> testRecovery(org.apache.jackrabbit.oak.plugins.segment.file.FileStoreIT)
>>>
>>>> Time elapsed: 0.554 sec  <<< ERROR!
>>>> java.io.IOException: Could not remove broken tar file
>>>> target\FileStoreIT1303808629179878423dir\data00000a.tar
>>>>          at
>>>>
>>>> org.apache.jackrabbit.oak.plugins.segment.file.TarReader.backupSafely(TarReader.java:237)
>>>>          at
>>>>
>>>> org.apache.jackrabbit.oak.plugins.segment.file.TarReader.collectFileEntries(TarReader.java:196)
>>>>          at
>>>>
>>>> org.apache.jackrabbit.oak.plugins.segment.file.TarReader.open(TarReader.java:125)
>>>>          at
>>>>
>>>> org.apache.jackrabbit.oak.plugins.segment.file.FileStore.<init>(FileStore.java:434)
>>>>
>>>>
>>> (yes, on Windows)
>>>
>>>
>> Is it failing consistently?
>>
>
> Yup, consistently. Manfred is seeing the same on Windows.
>
> As code and test probably did not change since yesterday -- did the
> re-packaging change anything about test execution?
>
> Best regards, Julian
>

Reply via email to