[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Debian needs more buildds. It has offers. They aren't being accepted.

Anthony Towns <aj@azure.humbug.org.au> writes:

> On Fri, Feb 13, 2004 at 10:36:19PM +0100, Ingo Juergensmann wrote:
> > On Fri, Feb 13, 2004 at 08:48:38PM +1000, Anthony Towns wrote:
> > > Generally, I think being frank about what you think -- like saying
> > > someone's acting like an idiot when you think they are -- is a good thing
> > > in technical discussions. I doubt you'll find anywhere that I've said
> > > otherwise. But when it crosses the line to constant complaints about
> > > people and suggestions that the project would be better off without
> > > their efforts, that's unacceptable, to my mind.
> > The discussion came rather quick to that point that that was only meant in
> > buildd related regards. 
> If this were the first such message ever, you'd have a point. See, eg,
> http://lists.debian.org/debian-devel/2004/debian-devel-200401/msg01629.html
> from the 19th of last month.

>From that mail:
Everything is in sid now, but the buildds FUCKING SUCK! The buildd
admins must be incompetent or on crack. They failed all the builds that
failed due to the g++ RC enum bug instead of updating their buildd and
setting the packages back to Needs-Build. The g++ RC enum bug was already
fixed before I even uploaded the packages which shows you how slow they
are to fix problems on their machines... However, people like IJ are

[IJ is not a DD and doesn't and shouldn't fix the buildd unless its an

refused to help maintain buildds because they have too little experience,

BTW - The override for kdeutils still needs to be done.

He is partly right complaining about failing packages that fail due to
an RC bug in the toolchain (or not yet compiled packags). Its partly a
problem in the buildd/wanna-build implementation and the buildd admins
job to keep an eye on. In the case of build-essential packages a
Dep-Wait will also not work right since sbuild does not update
installed packages unless the source needs it.

I hope Chris learned his lesson and subsequent mails have a much nicer
tone and show more understanding of the buildd system. He got replies
form the people involved in the process and learned from his mistakes
(we will see if it lasts or if he forgets again). Thats how it should
work. Just ignoring someone doesn't teach anything.

There is no way to have a package Dep-Wait on a bug, which would be
the real solution to the problem. In most cases I think setting a
Dep-Wait to the next version of the faulty package is the best
solution, the RC bug will probably be fixed by then.

Even more stupid is setting packages with missing Build-Depends to
failed instead of Dep-Wait (yes, that still happens).


Reply to: