New initscripts released, here you get the git shortlog:

Aaron Griffin (1):
Remove the kernel 'quiet' handling for fsck

Petar Bogdanovic (1):
Allow the user to cancel an fsck

Thomas Bächler (2):
Fix bug #13205: better random-seed entropy - For the random-seed
file location to be more "FHS compliant" is better to put in
/var/lib/misc/random-seed (Variable state information, that persists
after reboot) than the /var/run/random-seed (Run-time variable data,
that not necessarily persists after reboot) for example with /var/run
mounted as TMPFS. - The random-seed file permision should be "0600" not
"0644" - The size of random-seed acording to random(4) manpage should be
the size specified in "/proc/sys/kernel/random/poolsize" in other case
512 bytes.
Use the device nodes from /lib/udev/devices/* instead of
hardcoding them in initscripts. This requires a new udev package (at
least 139-1)


Tested on x86_64, please sign off.

On Sat, Mar 7, 2009 at 8:10 PM, Thomas Bächler <thomas@archlinux.org> wrote:
> New initscripts released, here you get the git shortlog:
>
> Aaron Griffin (1):
> * * *Remove the kernel 'quiet' handling for fsck
>
> Petar Bogdanovic (1):
> * * *Allow the user to cancel an fsck
>
> Thomas Bächler (2):
> * * *Fix bug #13205: better random-seed entropy - For the random-seed file
> location to be more "FHS compliant" is better to put in
> /var/lib/misc/random-seed (Variable state * information, that persists after
> reboot) than the /var/run/random-seed (Run-time variable data, that not
> necessarily persists * after reboot) for example with /var/run mounted as
> TMPFS. - The random-seed file permision should be "0600" not "0644" - The
> size of random-seed acording to random(4) manpage should be the size
> specified in "/proc/sys/kernel/random/poolsize" in * other case 512 bytes.
> * * *Use the device nodes from /lib/udev/devices/* instead of hardcoding
> them in initscripts. This requires a new udev package (at least 139-1)
>
> Tested on x86_64, please sign off.
>
>

Systems booted fine. Signoff both arches.

Eric

On Sun, Mar 8, 2009 at 3:51 PM, Eric Bélanger <snowmaniscool@gmail.com> wrote:
> On Sat, Mar 7, 2009 at 8:10 PM, Thomas Bächler <thomas@archlinux.org> wrote:
>> New initscripts released, here you get the git shortlog:
>>
>> Aaron Griffin (1):
>> * * *Remove the kernel 'quiet' handling for fsck
>>
>> Petar Bogdanovic (1):
>> * * *Allow the user to cancel an fsck
>>
>> Thomas Bächler (2):
>> * * *Fix bug #13205: better random-seed entropy - For the random-seed file
>> location to be more "FHS compliant" is better to put in
>> /var/lib/misc/random-seed (Variable state * information, that persists after
>> reboot) than the /var/run/random-seed (Run-time variable data, that not
>> necessarily persists * after reboot) for example with /var/run mounted as
>> TMPFS. - The random-seed file permision should be "0600" not "0644" - The
>> size of random-seed acording to random(4) manpage should be the size
>> specified in "/proc/sys/kernel/random/poolsize" in * other case 512 bytes.
>> * * *Use the device nodes from /lib/udev/devices/* instead of hardcoding
>> them in initscripts. This requires a new udev package (at least 139-1)
>>
>> Tested on x86_64, please sign off.
>>
>>
>
> Systems booted fine. Signoff both arches.

Another signoff for x86_64 here.

-Dan

Dan McGee schrieb:

Systems booted fine. Signoff both arches.


Another signoff for x86_64 here.


I'll take your signoff as signoffs for udev too (this initscripts
release depends on the new udev), but I'd love to have signoffs for
klibc-udev as well.

Thomas Bächler schrieb:

Tested on x86_64, please sign off.



Can we have another signoff on i686 for udev+klibc-udev+initscripts? I
want to move this soon.

Thomas Bächler wrote:

Thomas Bächler schrieb:

Tested on x86_64, please sign off.



Can we have another signoff on i686 for udev+klibc-udev+initscripts? I
want to move this soon.




My laptop still boots nicely.... sign-off i686.

Allan

> Thomas Bächler (2):
> * * *Fix bug #13205: better random-seed entropy - For the random-seed file
> location to be more "FHS compliant" is better to put in
> /var/lib/misc/random-seed (Variable state * information, that persists after
> reboot) than the /var/run/random-seed (Run-time variable data, that not
> necessarily persists * after reboot) for example with /var/run mounted as
> TMPFS. - The random-seed file permision should be "0600" not "0644" - The
> size of random-seed acording to random(4) manpage should be the size
> specified in "/proc/sys/kernel/random/poolsize" in * other case 512 bytes.
> * * *Use the device nodes from /lib/udev/devices/* instead of hardcoding
> them in initscripts. This requires a new udev package (at least 139-1)

I tested this on a minimal ArchLinux I installed in a virtual machine.
/var/lib/misc/ was not present on that install so rc.shutdown
complained, and the random-seed file of course wasn't created.



--
damjan

Damjan Georgievski schrieb:

I tested this on a minimal ArchLinux I installed in a virtual machine.
/var/lib/misc/ was not present on that install so rc.shutdown
complained, and the random-seed file of course wasn't created.


Uh, that needs to be fixed. initscripts is in core already, so you were
a little slow, but I'll try to push a fixed version tonight.