| 1 |
On Tue, 18 Dec 2012 22:05:21 -0500, Michael Mol wrote: |
| 2 |
|
| 3 |
> > I don't use separate initr* files, the initramfs is built into the |
| 4 |
> > kernel, using the latest versions of the tools installed at the time |
| 5 |
> > the kernel was compiled. That gives a single bootable file that, if |
| 6 |
> > it works now, should always work. Most changes to the component |
| 7 |
> > packages do not affect the simple job they have to do to get a system |
| 8 |
> > ready to run init. |
| 9 |
> |
| 10 |
> Just because it runs, doesn't mean it "works." For example, let's say |
| 11 |
> your network now requires an additional protocol for the host to |
| 12 |
> operate properly. Or let's say there's a security vulnerability in |
| 13 |
> some network-aware component in your initramfs. Or that some piece of |
| 14 |
> automounter code is vulnerable to corrupted filesystems on flash |
| 15 |
> drives. |
| 16 |
|
| 17 |
Those are unreasonable, if somewhat unlikely, scenarios. It would be |
| 18 |
simple to add a post_install hook that checks if a package used by |
| 19 |
your initramfs has been updated and uses einfo to advise you to rebuild |
| 20 |
your initramfs/kernel. |
| 21 |
|
| 22 |
|
| 23 |
-- |
| 24 |
Neil Bothwick |
| 25 |
|
| 26 |
Man and mouse are alike, both end up in pussy :) |
Archives