Re: A Suggestion for Release Testing

[René Rebe <rene at exactcode dot de>]

Hi,

On Sunday 12 June 2005 19:51, Gerald Pfeifer wrote:

> > Given the recent problems with the 4.0.0 release and major packages like
> > KDE and the kernel, has anyone considered testing releases by completely
> > compiling a Linux system?

> > I'm willing to implement this, if it's deemed useful.

Note that the T2 System Development Environmetn (http://www.t2-project.org) 
does already implement a build system with about 1600 package descriptions to 
build and bootrap a system including cross compilation and integrated distcc 
(and ccache - though this is less interesting in this context) support.

> > The downside is that the compile would take quite a while; even for my
> > dual Opteron, full builds take more than a day. On the other hand, this
> > might catch some problems before release.
> >
> > I'm willing to implement this, if it's deemed useful.
>
> This is certainly an excellent idea.  The more testing we can get,
> the better.

Bootstrapping a, reduced but including KDE, Linux desktop takes slightly more 
than a day on a, not that recent, Athlon - with OpenOffice about 7 hours 
longer. Bootrsapping all the 1600++ packages of T2 would take something in 
the range of over 4 days. Of course less on a dual opteron.

We already have contributor hosted regression testers running mostly 
permanently:

  http://t2.geeks.cl/regressions/stable/regressions.html

And the official T2 site will get a regression tester with permanent historic 
build erros and nice graph plots again soon, too.

Needless to say we support most of the Linux supported architectures to 
possibly test build on.

PS: Hardware sponsors welcome ,-)

Sincerely yours,

-- 
René Rebe - Rubensstr. 64 - 12157 Berlin (Europe / Germany)
            http://www.exactcode.de | http://www.t2-project.org
            +49 (0)30  255 897 45

Attachment: pgp00000.pgp
Description: PGP signature