@Bruno: multiple tests are using the same bucket and each test assume it is
resetted between them which is an assumption which is generally false.

Romain Manni-Bucau
@rmannibucau <https://twitter.com/rmannibucau> |  Blog
<https://rmannibucau.metawerx.net/> | Old Blog
<http://rmannibucau.wordpress.com> | Github <https://github.com/rmannibucau> |
LinkedIn <https://www.linkedin.com/in/rmannibucau> | Book
<https://www.packtpub.com/application-development/java-ee-8-high-performance>


Le mar. 18 déc. 2018 à 16:09, Bruno Baptista <[email protected]> a écrit :

> Romain, That issue does not seem to affect the failing tests.
>
> Can you please elaborate the problem with the 2 CircuitBreaker tests?
> Bruno Baptista
> https://twitter.com/brunobat_
>
>
> On 18/12/18 14:27, Bruno Baptista wrote:
>
> Thanks
> Bruno Baptista
> https://twitter.com/brunobat_
>
>
> On 18/12/18 14:25, Romain Manni-Bucau wrote:
>
> The one you are in charge ;):
> https://github.com/eclipse/microprofile-fault-tolerance/issues/355
>
> Romain Manni-Bucau
> @rmannibucau <https://twitter.com/rmannibucau> |  Blog
> <https://rmannibucau.metawerx.net/> | Old Blog
> <http://rmannibucau.wordpress.com> | Github
> <https://github.com/rmannibucau> | LinkedIn
> <https://www.linkedin.com/in/rmannibucau> | Book
> <https://www.packtpub.com/application-development/java-ee-8-high-performance>
>
>
> Le mar. 18 déc. 2018 à 15:11, Bruno Baptista <[email protected]> a
> écrit :
>
>> Do you have a link?
>> Bruno Baptista
>> https://twitter.com/brunobat_
>>
>>
>> On 18/12/18 14:08, Romain Manni-Bucau wrote:
>>
>> Yes
>>
>> Romain Manni-Bucau
>> @rmannibucau <https://twitter.com/rmannibucau> |  Blog
>> <https://rmannibucau.metawerx.net/> | Old Blog
>> <http://rmannibucau.wordpress.com> | Github
>> <https://github.com/rmannibucau> | LinkedIn
>> <https://www.linkedin.com/in/rmannibucau> | Book
>> <https://www.packtpub.com/application-development/java-ee-8-high-performance>
>>
>>
>> Le mar. 18 déc. 2018 à 14:10, Bruno Baptista <[email protected]> a
>> écrit :
>>
>>> Is this a know bug registered somewhere?
>>> Bruno Baptista
>>> https://twitter.com/brunobat_
>>>
>>>
>>> On 18/12/18 13:02, Romain Manni-Bucau wrote:
>>>
>>> No, to the wrong assumption in terms if CDI exception in terms of error
>>> during the deployment. The TCK felt in an unspecified area but we applied
>>> the same workaround we have for CDI tck for both potential exception in
>>> 2.0.9.
>>> Other workarounds, as the circuit breaker bug in the TCK,  are in scope
>>> tests so we are good.
>>>
>>> Romain Manni-Bucau
>>> @rmannibucau <https://twitter.com/rmannibucau> |  Blog
>>> <https://rmannibucau.metawerx.net/> | Old Blog
>>> <http://rmannibucau.wordpress.com> | Github
>>> <https://github.com/rmannibucau> | LinkedIn
>>> <https://www.linkedin.com/in/rmannibucau> | Book
>>> <https://www.packtpub.com/application-development/java-ee-8-high-performance>
>>>
>>>
>>> Le mar. 18 déc. 2018 à 13:56, Bruno Baptista <[email protected]> a
>>> écrit :
>>>
>>>> Was it related to this fix?
>>>>
>>>> https://github.com/eclipse/microprofile-fault-tolerance/issues/323
>>>>
>>>> Cheers
>>>> Bruno Baptista
>>>> https://twitter.com/brunobat_
>>>>
>>>>
>>>> On 18/12/18 12:53, Bruno Baptista wrote:
>>>>
>>>> Hi Romain,
>>>>
>>>> I have concerns about tweaking the test environment in order to pass
>>>> the TCK:
>>>>
>>>>
>>>> https://github.com/apache/geronimo-safeguard/blob/master/safeguard-impl/src/test/java/org/apache/safeguard/impl/tck/SafeguardTCKExtension.java#L80
>>>>
>>>> Without that, we have these failures:
>>>>
>>>> [ERROR] Failures:
>>>> [ERROR]
>>>> CircuitBreakerTest>Arquillian.run:138->testCircuitInitialSuccessDefaultSuccessThreshold:292
>>>> serviceA should throw an Exception in testCircuitDefaultSuccessThreshold on
>>>> iteration 5
>>>> [ERROR]
>>>> CircuitBreakerTest>Arquillian.run:138->testCircuitLateSuccessDefaultSuccessThreshold:370
>>>> in serviceA no CircuitBreakerOpenException should be fired on iteration 1
>>>> [INFO]
>>>> [ERROR] Tests run: 212, Failures: 2, Errors: 0, Skipped: 0
>>>>
>>>> You mention that the TCK leaks, can you provide further details on the
>>>> nature of the leak and how the TCK can be improved? I don't see a filed
>>>> issue regarding this problem.
>>>>
>>>> Cheers
>>>> Bruno Baptista
>>>> https://twitter.com/brunobat_
>>>>
>>>>
>>>> On 18/12/18 09:42, Romain Manni-Bucau wrote:
>>>>
>>>> Hi guys,
>>>>
>>>> as mentionned here is the vote fore Geronimo Safeguard 1.2.0
>>>>
>>>> The staging repo is:
>>>> https://repository.apache.org/content/repositories/orgapachegeronimo-1
>>>> <https://repository.apache.org/content/repositories/orgapachegeronimo-1073>
>>>> 075
>>>> My keys is still available in http://svn.apache.org/repos/asf/geronimo
>>>> /KEYS
>>>> Tag is on the main source repo and sources in the staging repo.
>>>>
>>>> This vote is open for 3 days as usual or untll it gets its 3
>>>> binding +1s.
>>>>
>>>> Romain Manni-Bucau
>>>> @rmannibucau <https://twitter.com/rmannibucau> |  Blog
>>>> <https://rmannibucau.metawerx.net/> | Old Blog
>>>> <http://rmannibucau.wordpress.com> | Github
>>>> <https://github.com/rmannibucau> | LinkedIn
>>>> <https://www.linkedin.com/in/rmannibucau> | Book
>>>> <https://www.packtpub.com/application-development/java-ee-8-high-performance>
>>>>
>>>>

Reply via email to