How to Classify Intermittent Test Failures

[Gedare Bloom]

On Mon, Feb 1, 2021 at 10:31 AM Joel Sherrill <joel at rtems.org> wrote:

>
>
> On Mon, Feb 1, 2021 at 11:14 AM Gedare Bloom <gedare at rtems.org> wrote:
>
>>
>>
>> On Mon, Feb 1, 2021 at 9:42 AM Joel Sherrill <joel at rtems.org> wrote:
>>
>>> Hi
>>>
>>> On the aarch64 qemu testing, we are seeing some tests which seem to pass
>>> most of the time but fail intermittently. It appears to be based somewhat
>>> on host load but there may be other factors.
>>>
>>>
>> There does not appear to be a good test results state for these. Marking
>>> them expected pass or fail means they will get flagged incorrectly
>>> sometimes.
>>>
>>>
>> Are they expected to pass every time?
>>
>
> Normally yes. But there appears to be some external factors particularly
> load impacting qemu which lead to failures on target side code.
>
>
Since they can be made to pass by reducing the load (I assume), then you
should document this in the target documentation and not change the testing
expectations. The test should pass. If too much load cause qemu to fail
randomly, this should be known and avoided or else your test results are
much less useful.

Does anyone know this happens on other architectures with qemu?


>
>> Intermittent failures are suspicious, and there is limited value to
>> running a test that has an "intermittent failure" state, since it will
>> always be successful if you don't care if it passes or fails.
>>
>
> Yeah. No matter what you mark them, it sucks.
>
>>
>>
>>
>>> I don't see not running them as a good option. Beyond adding a new state
>>> to reflect this oddity, any suggestions?
>>>
>>> --joel
>>> _______________________________________________
>>> devel mailing list
>>> devel at rtems.org
>>> http://lists.rtems.org/mailman/listinfo/devel
>>
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.rtems.org/pipermail/devel/attachments/20210201/345ca64d/attachment.html>