[lsb-discuss] False positive FAIL

Stew Benedict stewbintn at gmail.com
Thu Jan 5 13:41:33 UTC 2012


On 01/05/2012 12:40 AM, Xie Ruohong wrote:
> Hello,
>
> We developed an operating system and hope to get LSB certification. We 
> have run LSB distribution checker and there are some fails which we 
> believe are due to the incorrect tests.
>
> The four fails are list below:
>
> */olver/time/timer/tests/timer_getoverrun_spec 1failed*
> Message from the test:
> failure 285
> --------------------------------
> Scenario time_timer_created_timers_scenario:
> MEDIATOR_FAILED
> Execution failed
> Comment:
> Test Purpose Comment: Check extra requirement
> Open the journal:
> /var/opt/lsb/test/manager/results/IA64-sp3k-2011-12-30-16h-28m-02s/results/olver.journal:23602
>
> */olver/time/timer/tests/timer_getoverrun_spec 2failed*
> Message from the test:
> failure 286
> --------------------------------
> Scenario time_timer_timers_crash_scenario:
> MEDIATOR_FAILED
> Execution failed
> Comment:
> Test Purpose Comment: Check extra requirement
> Open the journal:
> /var/opt/lsb/test/manager/results/IA64-sp3k-2011-12-30-16h-28m-02s/results/olver.journal:23611
>
> */olver/time/timer/tests/timer_gettime_spec 1failed*
> Message from the test:
> failure 287
> --------------------------------
> Scenario time_timer_timers_crash_scenario:
> MEDIATOR_FAILED
> Execution failed
> Comment:
> Test Purpose Comment: Check extra requirement
> Open the journal:
> /var/opt/lsb/test/manager/results/IA64-sp3k-2011-12-30-16h-28m-02s/results/olver.journal:23670
>
> */olver/time/timer/tests/timer_delete_spec 1failed*
> Message from the test:
> failure 288
> --------------------------------
> Scenario time_timer_timers_crash_scenario:
> MEDIATOR_FAILED
> Execution failed
> Comment:
> Test Purpose Comment: Check extra requirement
> Open the journal:
> /var/opt/lsb/test/manager/results/IA64-sp3k-2011-12-30-16h-28m-02s/results/olver.journal:23788
>
> The four fails have similar output. I check the test source code and 
> found they use incorrect timerid. In these tests, a timer will be 
> created by /timer_create()/ and return /timerid/. The program use 
> /int/ type to store /timerid/. But /timerid/ is 64-bits while /int/ is 
> 32-bits on IA64 platform. So the/timerid/ will be truncated to an 
> incorrect value. Then test program will use the incorrect value and 
> generate a segmentation fault.
>
> How should we deal with these fails?
>
Please file a bug at lsbbugs.linuxfoundation.org. Looking at the open 
bugs on olver-core I don't see anything that looks similar to this. In 
the past we have run validation runs of the tests, pre-release on all 
the architectures and I generally open bugs on issues such as this that 
seem arch specific. At the moment I don't have access to an IA64 machine 
to check.

Opening a bug will be a starting point for any possible waiver and an 
entry in the problem_db (when it comes back online).

You can still submit your results for certification with the failures if 
you're convinced it's a test issue and it will be addressing during your 
results audit.

-- 

Stew Benedict


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linuxfoundation.org/pipermail/lsb-discuss/attachments/20120105/96b5e5e9/attachment.html>


More information about the lsb-discuss mailing list