[postgis-devel] Interrupt Test considered Harmful

Javier Santana jsantana at cartodb.com
Thu Jul 30 00:45:05 PDT 2015


Hi,

I think some time should spend on fixing those tests, we can do it for sure
but is there a way to test it properly, like moving them to the C tests
where we can have more control ?



On Thu, Jul 30, 2015 at 12:38 AM, Sandro Santilli <strk at keybit.net> wrote:

> On Wed, Jul 22, 2015 at 08:34:33AM -0700, Paul Ramsey wrote:
> > I’m going to move the interrupt test out of the main ‘check’ path, so
> that continuous integration will stop providing so many false positives.
> I’m getting close to ignoring debbie now because every second commit tosses
> an error, either from interrupt or because of one of failed cleanup issues
> that seems to randomly happen. CI is only useful if it’s actually returning
> good results, otherwise it’s just spam and gets treated as such.
> >
> > Anyways: moving interrupt out of ‘check’ into ‘checkall’ or something
>
> Pity, but I understand the problem.
> It would be nice to be able to use CPU time rather than wall time
> for checking that the interruption happens at "predictable" time,
> maybe you know of an SQL-exposed function to fetch that time ?
> (spent on the backend, not frontend).
>
> --strk;
> _______________________________________________
> postgis-devel mailing list
> postgis-devel at lists.osgeo.org
> http://lists.osgeo.org/cgi-bin/mailman/listinfo/postgis-devel




-- 
www.cartodb.com · CTO
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/postgis-devel/attachments/20150730/65fb4457/attachment.html>


More information about the postgis-devel mailing list