Bug#680693: unblock: qemu-kvm/1.1.0+dfsg-1
On 08.07.2012 23:10, Cyril Brulebois wrote:
> Michael Tokarev <mjt@tls.msk.ru> (08/07/2012):
>> That version grew a bug which people were hitting often -- #679788,
>> thats the only reason I hurried up with the new version. People
>> started upping severity exactly to STOP it from entering testing.
>> So there's no way it'd enter testing with this bug, even it if
>> has an easy workaround.
>
> Heh? It's your call as the maintainer of a given package to set a bug
> report's severity. Depending on the actual bug and the easiness of the
> workaround, sticking to severity: important could have worked just fine.
As stated in #679788, in message #29, I'm trying to value users,
and I agreed this should not enter testing. If something is hurting,
I agree it is better to fix it before letting it entering -testing.
I just didn't expect it to cause complications...
> Of course, we could have answered that if you had actually asked us
> before uploading and killing the automatically-granted exception…
..because I shuold have asked first.
>> What I can say for sure is that I wont be able to support 1.0 version
>> (which is currently in testing). It will be difficult already to
>> support 1.1, due to the same reasons, but at least the bugs already
>> fixed in 1.1 are not needed to be backported.
>
> Are you saying we should never release qemu-kvm? Or that you /could/
> work out something if we would let 1.1 in wheezy?
I'm trying my best here. Not having qemu-kvm at this stage is
a large miss. But having even older version to start with just
makes things much more difficult than they already are. So no,
I'm not saying we should never release it (especially since we
ship 0.12 version in squeeze currently -- Debian is the last
distribution out there whcih still ships such an old version :),
I'm saying it will be much more difficult for me to roll back
to 1.0 at this stage. And it will need quite a few bugfixes
too (and hence freeze exceptions), if we'll actually find ways
to backport some fixes (closed by 1.1 and a few more which I
know about) in a reliable way.
>> Well. Let's remove it from wheezy when. I can work it out in the
>> bpo, it will even be easier this way. After all, I already carried
>> it for about 2 years from my site during lenny times.
>
> Maybe we could make something happen anyway, e.g. by letting it stay for
> a longer period in unstable before considering an unblock. But I'll let
> some other team members voice their opinion on this topic.
Fair point.
Thank you!
/mjt
Reply to: