On Sun, Jul 29, 2007 at 10:38:18AM +0200, Raphael Hertzog wrote: > On Sat, 28 Jul 2007, Andreas Barth wrote: > > And, BTW, the buildd admins of the experimental buildds are in touch > > with the buildd admins of the unstable buildds - and I discussed that > > matter with Ryan and James before setting up the first buildds. Now you > > might see the difference: Because in the buildd case, it makes sense to > > distribute the load on more shoulders (and btw, it doesn't buy real > > advantages to force the same people handling unstable to also care about > > experimental). > > Agreed it makes sense to distribute the load on more shoulders. It doesn't > make sense to do it on non .d.o machines and it doesn't make sense to have > two wanna-build instances.
It might help if you knew how wanna-build and buildd worked. Since experimental building requires a few patches to sbuild, it doesn't make much sense to experimental autobuilding on hosts that also do unstable autobuilding. It might make sense to make that second host a d.o machine, too, but it doesn't matter all that much IMO. Since wanna-build requires a separate database for each and every distribution it tracks, (e.g., wanna-build.d.o has a database for oldstable, oldstable-security, stable, stable-security, testing, and unstable for each architecture it tracks) it doesn't matter much whether you run it on buildd.d.o or on another host -- you need a second instance anyway. [...] -- <Lo-lan-do> Home is where you have to wash the dishes. -- #debian-devel, Freenode, 2004-09-22 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
