On Sáb 11 Aug 2012 12:18:22 Adam D. Barratt escribió: > On Sat, 2012-08-11 at 22:43 +0800, YunQiang Su wrote: > > I update my system (Sid) just mow, and got some errors due to BinNUM of > > QT. [snip] > > The problem is there unsloved, and USERs are using my system, and > > We know exactly that the binnum of some packages will broken system > > certainly. Hi YunQiang! I was the maintainer who requested the binNMU because of a bug in GCC which caused problems in apps like okular. To tell you you truth, I wasn't aware of the possibility of causing another bug. On the other hand, if I had known, I would have also asked the binNMU because of the annoyance of the bug gcc caused (and I'm fairly sure that it could have triggered another bugs too, they just weren't discovered). > If you're running a multi-user system and basing it on sid, you really > have to expect some breakage, multi-arch or not. Adam is totally right here. If you need a multi-user system please consider avoiding testing/sid. Else I'm afraid you are too exposed to this kinds of bugs. > > Why we still BinNMU these packages before this problem completely > > resolved? > > > > Must we? > > Well it avoids tying up several hours of buildd time on architectures > not affected by the issue the binNMU was scheduled for (i.e. 9/14). Indeed, as Adam says, uploading Qt needs a lot of buildd time (apart from the time it takes to build + upload for the maintainer itself). We (as in pkg-kde team) have decided to avoid reuploading Qt for some more time in case some other bug appears. On the other hand, I really don't know if this kind of upload would be acceptable by the RT for unblock. If you need to solve this bug, feel free to rebuild Qt for yourself (consider using cowbuilder or sbuild). Kinds regards, Lisandro. -- Vió, buteó y andó Lisandro Damián Nicanor Pérez Meyer http://perezmeyer.com.ar/ http://perezmeyer.blogspot.com/
Attachment:
signature.asc
Description: This is a digitally signed message part.