| Age | Commit message (Collapse) | Author |
|---|
|
Debian never shipped the PS3PF legacy plattform,
only retain the relevant upstream accepted ps3 plattform drivers.
|
|
modules added to MODULES=most
the net section was missing quite a lot of the newer net driver,
add them all per hand. as we can't just ship drivers/net.
|
|
purpose of function changed, now we keep the old initramfs
on the real path until mkinitramfs run finished.
|
|
Attached is a patch to improve the way update-initramfs handles the
generating of new or updated initramfs images for kernels. It puts in
place a few measures to ensure that if there is not enough disk space on
/boot, a previous initramfs is still in tact, to allow the booting of
that kernel. The patch applies against current git head. See the Ubuntu
specification found at the following URL for more information:
https://wiki.ubuntu.com/HardyInitramfsErrorHandling
This patch does the following:
* When generating a new initramfs, instead of copying the old initramfs to a
backup file, it hard links it instead, so no more space is used than
necessary. it only copies to the backup file in the event that the filesystem
on /boot doesn't support hard links, eg FAT32 partitions.
* A new file is used to generate the initramfs. If the generation succeeds, it
is moved to the original initramfs's location, ready to use. If it fails, the
original initramfs is not affected, allowing it to be used at boot.
[ wrap on long line -maks ]
|
|
Parse cmdline for resume_offset, export it and pass it to
the klibc resume binary.
Based on a patch by Alan Jenkins <alan-jenkins@tuffmail.co.uk>.
Bonus small codingstyle clean up of local-premount/resume.
|
|
found on testing that configure_networking() shouldn't make
assumption that ${DEVICE} is really already set.
this still allows to bail out if device is passed and set.
|
|
This is a lightly modified version of Ian Jackson's original patch,
incorporating maximilian attems's comments.
dh_installdeb automatically installs the triggers file, so the rules file
does not need to be changed.
|
|
udev takes care to load the relevant modules.
also PS3PF never had landed in debian thus pretty useless
to take care of it just nuke it and be happy that
udev loading is fixed since 2.6.23-rc1.
|
|
based on a patch by Luke Yelavich <luke.yelavich@canonical.com>.
|
|
IPOPTS can be assumed to get passed :::::: and thus configure
any possible interface. we have thus also to source the relevant
ipconfig output :)
based on a patch by Michal Sojka <sojkam1@fel.cvut.cz>:
"If I want several computers to boot from the same ramdisk (with NFS
root) and some computers have multiple network interfaces, the DEVICE
variable can't be set to a specific value (e.g. eth0). Ipconfig from
klibc supports a mode, where DHCP request is sent to all interfaces
and the one receiving the first DHCP offer is used. This perfectly
suits my needs, but it can't be used with initramfs scripts because
the interface name in not known in advance and the DEVICE variable is
set to something like ::::::."
|
|
Previsous changes depend on the latest klibc,
thus bump dependency version.
|
|
No need to call chmod later on, now that klibc mknod can set permissions.
|
|
if /dev/console doesn't yet exist it should be created
with proper permissions.
|
|
klibc mkdir has only the short option, use that.
|
|
properly escape minus signs to make them easily pastable.
|
|
udev creates tty devices too, in the case of not using udev,
we need to ship more anyway, so no need to worry about that right now.
|
|
nuke fb device mknod creation as udev creates the fb device nodes.
suggested by waldi. positive test on qemu with usplash.
let's see if we get a bad interaction with usplash and vga=XXX boots.
|
|
These warnings are harmless and initramfs-tools wants to provide
this directory structure for other users.
scripts are to be placed there and dir existence is assumed.
|
|
Doubles the speed of mkinitramfs small size loss ~ 1%.
Suggested since some time by waldi and found in Ubuntu too now.
|
|
Those fields are official dpkg supported, so just use them.
|
|
be more consistent on exporting *all* relevant ROOT* variables,
thanks to Sten Spans <sten@blinkenlights.nl> for pointing that out.
|
|
This patch is part of three patches (initramfs-tools, cryptsetup,
dropbear) which enable mkinitramfs to create initramfs that provide the
ability to log in and unlock a cryptroot during the boot process from
remote via ssh.
Calling configure_networking from /scripts/functions might appear more
than once, so just try if it hasn't been done/wasn't successful yet.
Check that by testing for existence of /tmp/net-$DEVICE.conf which is
created by ipconfig.
|
|
|
|
Hi Kernel-Team,
i stumbled over this bug when I wanted to update my initrd while I had
some isos loop-mounted like this:
$ cat /proc/mounts |grep boot
/dev/hda1 /boot ext2 rw 0 0
/dev/loop57 /srv/pub/iso_loop/sys/bootrescue-cd-20071129-59 iso9660 ro 0
0
/dev/loop58 /srv/pub/iso_loop/sys/bootrescue-cd-dev iso9660 ro 0 0
/dev/loop59 /srv/pub/iso_loop/sys/bootrescue-dvd-20071129-59 iso9660 ro
0 0
/dev/loop60 /srv/pub/iso_loop/sys/bootrescue-dvd-dev iso9660 ro 0 0
The check for a ro-mounted /boot partition was falsly triggered by the
mount targets of the iso-images. As you can see in the patch attached,
for the conditional awk output to work as desired, another pair of hyphen
enclosing the awk match calls is required.
HTH,
Daniel Reichelt
|
|
|
|
|
|
|
|
it seems backport of current iniramfs-toosl together
with linux-images is more frequent then expected.
thus just declare that you need newer findutils then sarge available.
this can be kicked after lenny release.
|
|
Using mkinitramfs-kpkg to build the ramdisk.
awk: /^md/0/{print substr($5, 1, 3); exit}
awk: ^ syntax error
mkinitramfs-kpkg failed to create initrd image.
Failed to create initrd image.
/proc/mdstat naming unchanged, thuse use same awk parsing.
|
|
|
|
root nfs needs to check that the boot is not set before setting it.
Thanks Thomas Lange <lange@informatik.uni-koeln.de> for report!
|
|
Newer kernel in the effort of killing /etc/mtab show mount options,
like errors=remount-ro, that is matched for read only.
Thanks Jan Niehusmann <jan@gondor.com> for report.
|
|
Due to overly strict errexit usage by udev hook a lot might be missing.
This is an potential Etch -> Lenny upgrade issue. (closes: 431291)
fixes boot failure on UUID and LABEL root for those.
it seems much better to add this stupid hook than to pump udev dep.
Thanks Michael Prokop <mika@grml.org> for report.
|
|
|
|
check that the modalias really exists,
otherwise no need to add anything.
|
|
this allows to seperate conf files which needs to land on initramfs
and those that should only be run on mkinitramfs.
|
|
this is better then passing some strange prereqs param,
as the prereqs is intermixed with boot/hook scripts.
|
|
instead of searching non matching sysfs modules strings
PIIX_IDE != piix, just look up the modalias.
modprobe likes it too ;)
|
|
* missing i2o_block
* force_load when !udev
|
|
fstype just detects fine the fs,
even when running without udev.
|
|
do fstype stuff on top.
|
|
almost ready for release, just missing the hook-conf only dir.
|
|
If you have quiet in your /proc/cmdline you can never
reach any debug bootoption anymore.
|
|
You’ll notice a strange error message due to an empty
$init if the rootfs can’t be found.
|
|
fix dep_add_modules(),
as bonus check rootfs output so that eventual unkown
rootfs get reported.
|
|
thanks madduck for reporting the error:
[: 39: -q: unexpected operator
|
|
s/fbnum/fbno/
|
|
debian-live may as well be build on a host with ro /boot
|
|
ltsp in ubuntu started using NBD+unionfs+squashfs instead of NFS, and
debian-live uses has a network boot methods using cifs in addition to a
different way of using NFS (i think it also uses unionfs and maybe
squashfs, not just a plain NFS/cifs mount)...
so splitting out the networking related code into a separate function
would move towards not having forked code for all of these different
network boot methods. at least, that's my hope.
|
|
|
