Re: [Boost-testing] Completely out of date results...

From: Victor A. Wagner Jr. (vawjr_at_[hidden])
Date: 2005-05-01 17:43:33

At Sunday 2005-05-01 10:59, you wrote:
>On Sun, 01 May 2005 10:45:52 -0700, Victor A. Wagner Jr. wrote
> > At Sunday 2005-05-01 09:14, you wrote:
> > >Can we clean up the completely out of date results on the regression 
> board?
> > >Of course the best way to fix this would be for the regression testers 
> to post
> > >new results, but barring that, it would be nice to reduce the clutter 
> on the
> > >regression pages:
> >
> > I've suggested that we "grey out" (or "pastel out" if you prefer)
> >  the colors as the results age so that we don't have to look so
> > closely for ancient ones.
>
>My preference is to just eliminate them since the page is big enough as it is
>-- even on a big monitor...
>
> > I've also noticed, much to my dismay, that the timestamps given in,
> >  e.g. http://tinyurl.com/dgxwj have little, if anything, to do with
> > the actual time that the link, or test was run. Near as I can tell
> > they have something (I don't know what) to do with when
> > process_jam_logfiles (or whatever it's called) runs.
>
>Following the link I see the following timestamps:
>
>Report Time: Sun, 01 May 2005 16:15:18 +0000
>Linker output [2005-04-30 22:23:11 UTC]:
>Run output [2005-05-01 15:26:12 UTC]:
>
>Isn't it possible that the link step didn't rerun from a previous build and so
>only the run step was updated?  Otherwise I would agree that you would expect
>the link and run steps to be closer together.

oh, that was just a place that I knew had timestamps....
the previous evening I'd watched very closely as owi_mt_blahblah.exe ran 
(since I'd modified it so that it would complete in time.  I carefully 
marked the time it ran (I was watching the Task manager to see what 
resources were used also... it ran in the vicinity of 1958MST or 0258UT.... 
when the entire run was finished, I looked at the xml and it was reporting 
a timestamp of 0307UT.
I then did a grep on timestamp for the xml and they were all (those that 
were on that day) within 2 minutes of each other.  I'm sure we will all 
agree that even my wonderful machine isn't going to do all the links and 
runs in < 2 minutes.

in short, the times (for runs anyway) clearly have nothing to do with when 
they happen and it appears that they have a LOT to do with when the .xml 
file was created.

>Jeff
>_______________________________________________
>Boost-Testing mailing list
>Boost-Testing_at_[hidden]
>http://lists.boost.org/mailman/listinfo.cgi/boost-testing

Victor A. Wagner Jr.      http://rudbek.com
The five most dangerous words in the English language:
               "There oughta be a law"