|
|
Boost Testing : |
From: Anthony Williams (anthony_w.geo_at_[hidden])
Date: 2006-11-22 04:56:00
Vladimir Prus <ghost_at_[hidden]> writes:
> the current regression tables show a bunch of Python failures on
> JustSoftwareSolutions-V2, like this:
>
> http://tinyurl.com/y8j8u6
>
> the problem is that:
>
> 1. The results date is Nov 22.
> 2. The problem was fixed on Nov 18, and is gone from
> other testers.
>
> In fact, the linked page has this:
>
> Compiler output [2006-11-19 02:54:19 UTC]:
>
> so it looks like stale results are being submitted somehow.
Nope; it just takes a while for the tests to run. I started this run on Friday
evening (17th), and it wasn't finished until the early hours of Sunday morning
(19th).
The reason for the "22nd" date in the header is that the regression report
machine was stalled, so it hasn't processed the results until now.
> Any ideas? One possible explanation is that toolset name was changes, but
> results for the old toolset name were not cleared. If that's the case, they
> must be cleared manually.
I don't do incremental runs, so everything gets cleared every time.
Anthony
-- Anthony Williams Software Developer Just Software Solutions Ltd http://www.justsoftwaresolutions.co.uk