| 1 |
Dnia 2013-06-21, o godz. 10:16:10 |
| 2 |
William Hubbs <williamh@g.o> napisał(a): |
| 3 |
|
| 4 |
> On Fri, Jun 21, 2013 at 12:23:28PM +0200, Michał Górny wrote: |
| 5 |
> > > If eselect-init installs the wrapper as /sbin/einit, we don't have to |
| 6 |
> > > touch /sbin/init at all, then, the only thing someone would have to do |
| 7 |
> > > is to add an entry to their boot loader with init=/sbin/einit on the kcl |
| 8 |
> > > to use it. |
| 9 |
> > |
| 10 |
> > But *if* the wrapper fails to run somehow, e.g. becomes broken, |
| 11 |
> > the kernel will fallback to the standard location. |
| 12 |
> |
| 13 |
> Yes, but if the wrapper replaces /sbin/init, like it does now, and the |
| 14 |
> wrapper gets broken, I think you are left with an unbootable system. |
| 15 |
|
| 16 |
Then kernel falls back to safe /bin/sh which is a minimal safe fallback. |
| 17 |
|
| 18 |
-- |
| 19 |
Best regards, |
| 20 |
Michał Górny |
Archives