Re: Does it sometimes happen that people send mails before NMU ?
On Sun, Jan 15, 2006 at 11:31:10PM -0800, Steve Langasek <vorlon@debian.org> wrote:
> Hi Mike,
>
> On Mon, Jan 16, 2006 at 07:47:52AM +0100, Mike Hommey wrote:
> > There have been 2 NMUs on libxml2 in a week and I never got a message
> > beforehand. Now I wonder if that practice has disappeared somehow.
>
> > I admit I've not spent enough time for libxml2 recently, but still, I
> > wouldn't have been bothered by some poking beforehand.
>
> > Moreover, I'm not exactly sure the second NMU has indeed removed all
> > problematic content but the bug is closed, so the NMUer may be happy.
> > Ah, by the way, the bug was not even a problem for package propagation
> > to testing, so that doesn't make the propagation an argument for a quick
> > upload.
>
> The first bug shows a message from the NMUer apologizing for not sending his
> patch beforehand. The second bug shows a patch sent by the NMUer prior to
> the NMU. Did you not receive these mails?
The patch was 8 minutes prior to the NMU.
> > Moreover, I'm not exactly sure the second NMU has indeed removed all
> > problematic content but the bug is closed, so the NMUer may be happy.
> > Ah, by the way, the bug was not even a problem for package propagation
> > to testing, so that doesn't make the propagation an argument for a quick
> > upload.
>
> Well, no, but the fact that it's a longstanding release-critical bug, with
> no maintainer response, means that it does warrant NMUer attention. If some
> non-free files have been missed in the process, that would be bad, but that
> doesn't seem to be a reason to not try? Of course, it's your prerogative as
> maintainer to review the contents of the NMU for correctness before
> acknowledging the bug closure.
Yes, it's my fault i didn't tag the bug or sent any feedback, but I'm
actually trying to find a better solution than removing the files, with
upstream cooperation, considering that upstream adds new testcases quite
often, and that any addition is likely to have the same problem.
Fixing RC bugs just for the sake of making the RC bug count down is not
the best thing to do IMHO. Some might even only be important bugs tagged
as grave or serious for some wrong reasons.
Mike
Reply to: