[vtk-developers] Failures on murron and bubbles

[Sean McBride]

On Wed, 20 Mar 2013 14:07:01 -0400, David E DeMarle said:

>I like the idea of having the nightly expected being just for the
>particular platforms and dependency releases we explicitly support.
>
>I also like the fact that when submitters like these test dev head of our
>deps we know when there is trouble brewing.
>
>The problem we just hit is that it isn't obvious from a quick glance at the
>dashboard which test releases and which don't. A new dashboard section
>would make it obvious. The nightly section itself could fill that roll if
>we all agree now that that is what it is for and remember that from now
>on. Another problem is that the NIghtly section hasn't been maintained well
>enough for us to know if the dashboards in there _should_ be passing. As it
>is now we treat Nightly like an experimental section and for the most part
>ignore it.

My Rogue7 dashboards are also testing using a top-of-tree build of clang (though the clang build itself is not built nightly, just manually every few weeks).

This occasionally produces a little non-green if clang regresses in some way.

I also have the impression that it would be lost in noise if it was not in 'expected', since the 'nightly' section is not so green.

Cheers,

-- 
____________________________________________________________
Sean McBride, B. Eng                 sean at rogue-research.com
Rogue Research                        www.rogue-research.com 
Mac Software Developer              Montréal, Québec, Canada