Raphael Hertzog wrote...
> Hello Christoph,
>
> On Sun, 15 Feb 2015, Christoph Biedl wrote:
> > Nguyen Cong wrote...
> >
> > > I would like to send the debdiff file for file package.
> > > Could any one please review it and give me some comments.
> >
> > NACK. This does not fix the issue or introduces a new one.
>
> Can we please avoid "NACK" and use full sentences that convey the proper
> respect that people deserve when they try to contribute?
Certainly. See below.
> NACK is just rude.
It's a stern warning that something is going wrong here. Kept short
since the first thing is to avoid an upload that causes trouble, and
there was no information how much time there is left.
If you call that rude I don't know I should call the compliments I
have received for my attempts to improve Debian.
> > An updated file package for squeeze has already been prepared here and
> For the records, I don't see your name in
> https://anonscm.debian.org/viewvc/secure-testing/data/dla-needed.txt?view=markup
> which is where we indicate that we plan to take care of applying
> the relevant security fixes of a given package to squeeze-lts.
I indeed forgot about that this time. So go ahead.
> It's always nice when maintainers want to join the LTS effort, but
> most maintainers are not interested in taking care of their packages in
> squeeze-lts so it's quite common to release LTS updates without any
> coordination with the maintainer.
And that's a bad thing. The maintainer very likely has more experience
with the package than anybody else, and also with the packaging. Knows
the gotchas in the code. Has out-of-tree test suites, and better
connections to upstream.
Plus, when preparing up to four updates for an issue, a cross-
distribution comparion helps to identify possible shortcomings in the
fix.
Which in that case is: The output of file(1), before and after the
appplying the fix, changes with the provided reproducer. But does not
for squeeze, it's still ENOMEM after a huge allocation was tried.
Before you ask, that reproducer is linked in the bug report.
Christoph
Attachment:
signature.asc
Description: Digital signature