| 1 |
On Wed, Jan 23, 2013 at 7:32 AM, Fabio Erculiani <lxnay@g.o> wrote: |
| 2 |
> I hope this is going to be binary package manager friendly. |
| 3 |
> In Sabayon for instance, kernel sources are not even installed and at |
| 4 |
> the same time, /proc/config.gz may not even be available. |
| 5 |
> There were some corner cases in where pkg_setup failed because this |
| 6 |
> kernel config check stuff was trying to be "smarter" than the user. |
| 7 |
|
| 8 |
Another issue with this sort of thing is that the kernel in |
| 9 |
/usr/src/linux might not be the kernel that is running, and the kernel |
| 10 |
that is running might not be the kernel that is running upon the next |
| 11 |
boot. |
| 12 |
|
| 13 |
I think warnings just make sense, but if we're going to make them |
| 14 |
fatal then at least print out instructions on how to disable them. I |
| 15 |
suspect that a large number of users will end up disabling config |
| 16 |
checks though, in which case this feature will provide a false sense |
| 17 |
of security. |
| 18 |
|
| 19 |
If an upgrade could make a user's system unbootable then we should |
| 20 |
really be warning them about it BEFORE they merge the package. A news |
| 21 |
item before the change is committed to portage would be more |
| 22 |
appropriate. I don't want a fatal error during pkg_setup that isn't |
| 23 |
fatal for 90% of our users because of all the false alarms to be an |
| 24 |
excuse for developers to put the blame on the user. That amounts to |
| 25 |
telling our users that if they don't enable some make.conf option then |
| 26 |
half of the Gentoo devs will cause their builds to break randomly, and |
| 27 |
if they do enable it the other half of the Gentoo devs will cause |
| 28 |
their systems to not boot. |
| 29 |
|
| 30 |
Rich |
Archives