1 |
On Wed, Feb 10, 2016 at 09:52:29AM -0500, Rich Freeman wrote: |
2 |
> On Wed, Feb 10, 2016 at 9:27 AM, <waltdnes@××××××××.org> wrote: |
3 |
> > On Wed, Feb 10, 2016 at 12:09:58AM -0500, Mike Frysinger wrote |
4 |
> >> On 09 Feb 2016 22:39, Duncan wrote: |
5 |
> >> > Mike Frysinger posted on Tue, 09 Feb 2016 14:26:52 -0500 as excerpted: |
6 |
> >> > > On 08 Feb 2016 13:46, Micha?? Górny wrote: |
7 |
> >> > >> I'm strongly against this, because: |
8 |
> >> > > |
9 |
> >> > > agreed. i also don't see any reasons in Patrick's e-mail to suggest the |
10 |
> >> > > current default is inadequate. "i don't like upstream" isn't relevant. |
11 |
> >> > |
12 |
> >> > I'd agree, except that the way we're running udev is strongly discouraged |
13 |
> >> > and generally not supported by upstream, with a statement that it /will/ |
14 |
> >> > break in the future, it's simply a matter of time. |
15 |
> >> |
16 |
> >> start a thread then when that actually happens |
17 |
> > |
18 |
> > The problem with that approach is that all at once the Gentoo forum |
19 |
> > will be hit with questions by a whole bunch of people who will have to |
20 |
> > migrate to either eudev or systemd on a short deadline. As the old |
21 |
> > saying goes, an ounce of prevention is worth a pound of cure. I believe |
22 |
> > that the best way to handle a crisis is to prevent it in the first |
23 |
> > place. That means getting into a lifeboat before standalone udev sinks. |
24 |
> > |
25 |
> |
26 |
> What would force them to migrate all at once? Even if udev-229 |
27 |
> doesn't work standalone, udev-228 will work standalone forever, and |
28 |
> patches could always be backported if there were some security crisis. |
29 |
> |
30 |
> By all means change the virtual if it is the right thing to do, but I |
31 |
> don't think that we should treat this as some kind of risk that could |
32 |
> cause problems for us without warning. Most likely if a need to |
33 |
> abandon it comes up we'll have months of warning, with still-working |
34 |
> versions of udev not even marked stable yet in the pipeline. |
35 |
> |
36 |
> As I said earlier, the advantage of procrastination is that you can |
37 |
> make a choice later when you have better information or perhaps even |
38 |
> more choices available. Maybe by the time udev stops working |
39 |
> standalone everybody has seen the light and is now using libredev |
40 |
> anyway. |
41 |
> |
42 |
> Often the decision to procrastinate is a decision that is rewarded. |
43 |
> That should be considered carefully. |
44 |
|
45 |
+ 10000. |
46 |
|
47 |
I also saw another issue that made me shudder. If we change the default |
48 |
to eudev, people who are running separate /usr are going to think they |
49 |
can kill their initramfs's, because people in gentoo conflated the |
50 |
separate /usr and initramfs issue with udev [1]. |
51 |
|
52 |
William |
53 |
|
54 |
[1] https://bugs.gentoo.org/show_bug.cgi?id=573760#C4 |