4592b6983a
base: add logger for rdinitdebug
...
initlog logs to /init.log /dev/kmsg and /dev/console
2009-10-01 18:11:24 +02:00
a0e44f97bd
iscsi/iscsiroot: remove redirection on rdnetdebug
...
iscsiroot is now executed in the initqueue
2009-10-01 15:21:55 +02:00
de439e24d7
base/init: stop udev exec queue before killing it
2009-10-01 15:17:59 +02:00
187d04ccb0
syslog: use initqueue to start syslog
2009-10-01 14:46:16 +02:00
f214273fd6
nfs/nfsroot: do not redirect output on rdnetdebug
...
nfsroot is now in the initqueue
2009-10-01 14:46:16 +02:00
52aeb3f312
base/dracut-lib: add setdebug
...
do not clutter rdinitdebug output with getarg and getargs debug
2009-10-01 14:46:16 +02:00
d84c1df906
nbd/nbdroot: add better check if nbd0 is ready to be mounted
2009-10-01 14:46:15 +02:00
a1dd79d4fe
network/netroot: remove flock
...
netroot is now in the initqueue
2009-10-01 14:46:15 +02:00
0519556070
network/netroot: do not redirect output on rdnetdebug
...
netroot is now in the initqueue
2009-10-01 14:46:15 +02:00
d64d04df39
nbd/nbdroot: do not redirect debug output on rdnetdebug
...
nbdroot is now in the initqueue and rdinitdebug should cover that
2009-10-01 14:46:15 +02:00
cf476dbc41
nbd/nbdroot: use udevsettle to wait for /dev/nbd0
...
nbdroot is now in the initqueue, so we can use udevsettle
2009-10-01 14:46:15 +02:00
8bfed103d8
mdraid: do not cleanup on auto assembly with mdadm.conf
2009-10-01 14:46:14 +02:00
e9b6e4e839
nbd: finish on /dev/root rather than on nbd0
2009-10-01 14:46:14 +02:00
1dc53cfd1e
nfs: specify nfs version 3 explicitly and retry with nfs version 2
2009-10-01 14:46:13 +02:00
f57b5bc735
network: move netroot to initqueue
...
Schedule netroot for the initqueue inside ifup and dhclient.
2009-10-01 14:45:24 +02:00
0e0031dc61
init: start udevd with --resolve-names=never
2009-09-30 15:04:50 +02:00
2168327bfa
init: mount /dev/shm
2009-09-30 15:04:36 +02:00
99fd62e3be
dm/lvm: make usage of new udev event rules optionally
2009-09-29 18:29:34 +02:00
0cded44025
95dasd: add dasd_diag_mod module
2009-09-29 18:20:48 +02:00
ffe53c91c5
95dasd: use 'uname -m' instead of 'arch'
2009-09-29 17:10:22 +02:00
9a2860a3ee
95dasd: fixed typo in modules.conf generation
2009-09-29 14:51:21 +02:00
896199c652
95dasd: do not install zfcp kernel module
2009-09-29 14:48:06 +02:00
5fcc6b4a0a
rename 95s390 to 95dasd
2009-09-29 14:47:29 +02:00
678b3605e8
dmraid: add missing dso
2009-09-23 17:16:31 -07:00
b9c4654a63
devicemapper: add dmeventd and new rules for device mapper
2009-09-23 16:52:14 -07:00
845dd50275
Removing umount appears to be a mistake, it seems to have been OK prior.
2009-09-23 15:16:59 -04:00
23aa204856
Fix LiveCD boot regression caused by bf55e2c5d3
...
FIXME: There is still something wrong with this boot where it wants to use umount, but doing so breaks live boot.
2009-09-21 18:14:27 -04:00
8eb16b0827
selinux: bail out if policy could not be loaded and selinux=0 not
...
specified
2009-09-21 17:44:14 +02:00
fc1abcfa1f
dmsquash-live: add initqueue-finished
2009-09-19 10:51:34 +02:00
c53acc3015
lvm: add rd_NO_LVMCONF
2009-09-18 12:28:36 +02:00
7a34efa5c4
md/dm/lvm: install pre-pivot cleanup hooks only if actually needed
...
- removed dmraid cleanup
- only call lvm cleanup, if no lvm.conf provided
- copy over lvm.conf by default
- add --nolvmconf --lvmconf parameters
2009-09-18 12:20:21 +02:00
2727e45b43
kernel-modules: force install of scsi_wait_scan in hostonly mode
2009-09-18 12:20:04 +02:00
d6a5b04043
Recognize nodmraid and noiswmd kernel cmdline options
...
Recognize nodmraid and noiswmd kernel cmdline options, for anaconda /
rc.sysinit / mkinitrd compatibility.
2009-09-18 08:58:14 +02:00
1de9edc06e
Do not stop / start udev exec-queue around "mdadm -As --run"
...
When assembling containers + embedded arrays from mdadm.conf,
mdadm needs the /dev/md# node for the container to assemble the
arrays within the container. Stopping the udev exec queue, results in
this node not getting created and mdadm failing to online the
arrays within the container.
Not having stop / start udev exec-queue around "mdadm -As --run" should
be safe as the exact same command is run from rc.sysinit without
any queue locking.
2009-09-18 08:15:59 +02:00
0f7e10cef4
mdraid: save state dir for mdmon/mdadm for the real root
2009-09-17 12:13:54 +02:00
689f9b8af7
create /var/run in 99base
2009-09-17 12:10:48 +02:00
24bac034d1
65-md-incremental-imsm.rules: check if md raid is already running
2009-09-16 16:10:09 +02:00
094097fab3
lvm/dm/md: switch from lockfiles to $sys/$devpath/holders/*
...
This is a more sane solution, than ignoring subsequent "change" events.
The only danger is that we could loop, if a lvm scan triggers a broken
md partition, which triggers a broken PV and so on.
Better fix the scanning tools, not to emit change events for devices,
if no action was taken.
2009-09-16 15:34:13 +02:00
04a8f224c1
Add ifname= argument for persistent netdev names
...
ifname=<interface>:<MAC>
Assign network device name <interface> (ie eth0) to the NIC
with MAC <MAC>.
Note that if you use this option you *must* specify an ifname=
argument for all interfaces used in ip= or fcoe= arguments
2009-09-16 15:01:08 +02:00
597839331c
dmraid: use and install kpartx for dmraids
2009-09-16 11:21:24 +02:00
26b72485cf
90dmraid: use kpartx to scan for partitions
2009-09-16 11:20:57 +02:00
7dbaf59ccf
10rpmversion/install: fix non-rpm install
2009-09-16 11:18:22 +02:00
95cf0a8803
10rpmversion,dracut.spec: generate version file and install it
2009-09-16 10:43:50 +02:00
1dd5bc3fba
add udevproperty() to dracut-lib and use it
...
newer versions of udev use --property instead of --env
2009-09-16 10:43:50 +02:00
bf55e2c5d3
dmsquash-live-root requires umount
2009-09-15 17:17:57 -04:00
a52586e8ea
network: break udevsettle early
2009-09-15 15:38:53 +02:00
68f28d6976
Add ifname= argument for persistent netdev names
...
ifname=<interface>:<MAC>
Assign network device name <interface> (ie eth0) to the NIC with MAC <MAC>.
Note that if you use this option you *must* specify an ifname= argument
for all interfaces used in ip= or fcoe= arguments
2009-09-15 15:13:21 +02:00
5c3f5746bb
90mdraid/65-md-incremental-imsm.rules: first set initqueue-finished,
...
then initqueue-settled
also introduce md_finished.sh to check for all md activity to be settled
2009-09-15 15:00:25 +02:00
54a73cbb2b
dracut-lib.sh: change -f to -e to allow symlinks
2009-09-15 15:00:24 +02:00
ab0f04844b
10rpmversion: try to get the correct dracut rpm version during %post
2009-09-15 15:00:24 +02:00
3329f92136
90crypt/70-luks.rules: set unique name for initqueue job
2009-09-15 15:00:24 +02:00
c27ec060da
95rootfs-block/block-genrules.sh: beautify code
2009-09-15 15:00:24 +02:00
82b2189d01
initqueue-finished: fail also on empty queue
2009-09-15 15:00:24 +02:00
4ce52970c0
initqueue: first remove the job, then execute it
2009-09-15 15:00:24 +02:00
99433dd289
90mdraid: only switch mdmon to new root, if an instance is running
2009-09-15 15:00:23 +02:00
b94c520660
plymouth: add rd_NO_PLYMOUTH
2009-09-15 15:00:23 +02:00
2f02ae9d45
mdraid: add rd_NO_MDADMCONF --mdadmconf --nomdadmconf
...
Copy /etc/mdadm.conf to initramfs (even for non-hostonly) if
mdadmconf="yes" is set in dracut.conf or --mdadmconf is specified on the
dracut command line.
This was done, because there seems _no_ sane way to autoassemble md raid
arrays.
also moved rd_NO_MD to an udev ENV
2009-09-15 15:00:18 +02:00
11fcc8c8f9
95resume/install: give resume higher prio than mount
2009-09-15 11:17:28 +02:00
e8cf498dcc
95rootfs-block/mount-root.sh: ignore /.resume
2009-09-15 11:17:28 +02:00
2178a00eea
95resume: simplyfy rules and also recognize blkid typo
2009-09-15 11:17:27 +02:00
33ff830c1b
90dmraid: switch rd_NO_DM to udev ENV
2009-09-15 11:17:27 +02:00
5a8d8dfbb6
switch modules to initqueue-finished
2009-09-15 11:17:22 +02:00
8238850caa
init: add initqueue-finished
...
/initqueue-finished/*.sh are scripts, which return != 0, if the main
loop still has pending jobs to do.
2009-09-15 08:47:47 +02:00
1888e57bf9
plymouth: grep for PLYMOUTH_POPULATE_SOURCE_FUNCTIONS
2009-09-14 19:34:30 +02:00
4839a7995e
add pcmcia adapter
2009-09-14 19:30:20 +02:00
1aca526cb4
90dmraid: install parse-dm.sh
2009-09-14 16:47:30 +02:00
600131fe85
selinux-loadpolicy: do not check for selinux config files
...
leave it up to load_policy
2009-09-11 16:30:10 +02:00
8c45b484ce
plymouth: removed rd_plytheme
2009-09-11 16:29:40 +02:00
650da7bada
removed initrdargs from parse-*.sh
2009-09-10 17:34:15 +02:00
247c9f70bf
init: fixed command line handling for /sbin/init and init=<custom init>
2009-09-10 16:48:36 +02:00
233f79df99
selinux-loadpolicy.sh: also check for /etc/selinux/config
2009-09-10 15:49:03 +02:00
6f0e469d79
selinux-loadpolicy: set initial ret=0
2009-09-09 19:50:44 +02:00
f9749f368f
plymouth: use /usr/libexec/plymouth/plymouth-populate-initrd if possible
2009-09-09 18:05:00 +02:00
0b2de1f895
99base/init: output more info()
2009-09-09 18:04:55 +02:00
94eb3cf222
selinux-loadpolicy: do not mount /proc and /selinux
...
load_policy already does these things for us
2009-09-09 17:06:21 +02:00
cb47caf786
add parameter to specify additional kernel modules to include
...
--add-drivers for dracut
add_drivers for dracut.conf
2009-09-09 17:02:37 +02:00
d9f0633878
selinux-loadpolicy: do not mount /proc and /selinux
...
load_policy mounts it for us
2009-09-09 14:41:40 +02:00
71094bee63
base/init: add scsi_wait_scan to wait for host adapter scans
2009-09-09 11:09:04 +02:00
f5c4374d2c
selinux-loadpolicy.sh: mount /proc and /selinux before loading the
...
policies
also check for /sbin/load_policy
2009-09-08 16:26:37 +02:00
aa297ca6b6
65-md-incremental-imsm.rules: refine rules
2009-09-08 15:33:19 +02:00
4712e4ece6
initqueue: add --name parameter to give a name for --unique
2009-09-08 15:31:48 +02:00
f25c9697f8
mdraid: stop udev queue until mdraid assembled
2009-09-08 15:30:17 +02:00
7e282f5df1
dmraid: only install 64-md-raid.rules if no vol_id present
2009-09-08 15:29:30 +02:00
3bb5b261c7
65-md-incremental-imsm: add additional check for md-non-containers
2009-09-08 13:42:31 +02:00
273b197d75
added initqueue-settled and refactored code
2009-09-07 19:12:01 +02:00
906dedf4d5
mdraid-cleanup: do not stop mdraid container
2009-09-07 19:12:00 +02:00
4f4a79b1a1
add scp to debug module
2009-09-07 17:10:19 +02:00
587b3497c1
mdraid_start.sh: fixed raid activation
2009-09-07 16:18:12 +02:00
96f3cf1d27
95udev-rules: fixed c&p bug which did not install 61-persistent-storage.rules
2009-09-07 11:40:20 +02:00
e4dec3d6b6
{crypt,dmraid,lvm,mdraid}: cleanup with pre-pivot 30 and 31
2009-09-03 17:11:57 +02:00
429aa1a2c2
mdraid: add grep for convenience
2009-09-03 17:11:03 +02:00
52168b7b9d
reordered blkid/vol_id rules
2009-09-02 15:55:34 +02:00
d6462323d6
LVM/DM rules in dracut
...
I've looked at the LVM rules used in dracut just recently
and it needs fixing - we should react to change events only
for DM devices, so we have to skip vol_id/blkid call on ADD:
KERNEL=="dm-[0-9]*", ACTION=="add", GOTO="lvm_end"
Also, MD devices have their own rules, where vol_id/blkid
is called and where the symlinks are created (when looking
into raw initrd, this is in 64-md-raid.rules).
Also, if those rules are meant to be for DM devices only,
maybe we should skip symlink creation for the other devices
there, to keep the rules clean and straightforward. I think
we shouldn't create/recreate symlinks for non-dm devices in
LVM/DM rules (..should be in appropriate rules for that type
of device):
KERNEL!="dm-[0-9]*", GOTO="lvm_end"
2009-09-02 15:55:34 +02:00
1d368457e3
Merge commit 'origin/master'
2009-09-02 07:40:28 -05:00
7c3fe07ac0
Made hostonly checking for fcoe fail for now.
...
Having it unconditionally pass pulls in all the networking cruft even
for systems that do not need it, and that sorta defeats the purpose of
hostonly mode.
2009-09-02 07:37:33 -05:00
194f7d125f
90*/*cleanup.sh: fixed redirection to /dev/null
2009-09-01 15:59:31 +02:00
8ef0be5671
dracut-lib: try to set UDEVVERSION, if not set
2009-09-01 15:51:51 +02:00
b809e77f54
plymouth: in the generic case, only install text and details themes
2009-09-01 15:51:51 +02:00
780afa9876
release any raid/crypt/lvm device assembled, but not in use
2009-09-01 15:51:51 +02:00
5ebc1acb7f
selinux-loadpolicy.sh: redirect output of selinux load_policy
2009-09-01 15:51:50 +02:00
ec74fa8789
split dracut in network / non-network
2009-09-01 15:42:28 +02:00
c1d670a893
65-md-incremental-imsm.rules: also check on "change"
2009-09-01 15:42:27 +02:00
b6bf29b054
lvm.rules: blkid for cciss devices
2009-09-01 15:42:27 +02:00
5c6a593f2e
add udevsettle function and settle after raid/lvm/crypto activities
2009-09-01 15:42:27 +02:00
f4c36b28f6
mdraid_start: check if raid is already active before trying to run it
2009-09-01 15:42:27 +02:00
a89fb6daa5
Write netboot config for passing to OS as requested in https://bugzilla.redhat.com/show_bug.cgi?id=515771#c18
2009-08-31 19:34:27 -04:00
3508cc816e
Add fcoe booting support
...
Supported cmdline formats:
fcoe=<networkdevice>:<dcb|nodcb>
fcoe=<macaddress>:<dcb|nodcb>
Note currently only nodcb is supported, the dcb option is reserved for
future use.
Note letters in the macaddress must be lowercase!
Examples:
fcoe=eth0:nodcb
fcoe=4A:3F:4C:04:F8:D7:nodcb
2009-08-28 12:53:56 +02:00
84c2266f1a
network/installkernel: it's eth_type_trans and not eth_type_transfer
2009-08-27 13:06:46 +02:00
054e038512
Add accurate --hostonly checking for 95iscsi
2009-08-21 22:22:40 -05:00
240cc7c480
Abstract out hostonly vs. generic module installation differences.
...
This introduces filter_kernel_modules, which should be used to install
all kernel modules that match whatever criteria you want.
If running in --hostonly, filter_kernel_modules will only consider
modules that are loaded in the kernel, otherwise it will consider
all the modules installed on the system for the appropriate kernel.
This drastically reduces initramfs generation time when using --hostonly
by eliminating lots of unneeded filesystem activity.
2009-08-20 15:52:45 -05:00
16df64ef1f
Abstract out vol_id vs. blkid usage.
...
vol_id seems to be going away, and blkid can be used as a drop-in replacement
for it.
2009-08-20 07:28:04 -05:00
f76ef3aa3d
Speed up loading block drivers when running with --hostonly
...
Instead of grovelling through all the modules available for the
kernel looking for block devices, only look at the modules that are
actually loaded. This speeds things up by a rather large amount
when generating the initramfs with --hostonly.
While we are at it, only load the filesystem module that will actually
be used for the root filesystem when running in --hostonly instead
of all the filesystem modules that happen to be loaded at the time.
2009-08-19 22:21:29 -05:00
59c88f499f
Abstract out vol_id vs. blkid usage
...
Since different distros may or may not use vol_id in udev, and blkid
is generally replacing vol_id, abstract them out into a function which
tries to use vol_id first and blkid second, on the assumption that
blkid can take over for vol_id if vol_id is no longer there.
2009-08-19 22:21:07 -05:00
afe25d0ccc
{crypt,dmraid,,lvm,mdraid}/check use blkid if no vol_id found
2009-08-19 17:11:29 +02:00
495a964322
nbd/check: moved host-only checks and added ";"
2009-08-19 17:09:53 +02:00
68b26290f0
udev-rules: install vol_id
2009-08-19 11:12:42 +02:00
d41f7ae750
rsyslogd-start.sh: read correct filters file
2009-08-19 10:59:37 +02:00
e15bc1f330
Syslog support for dracut
...
This module provides syslog functionality in the initrd.
This is especially interesting when complex configuration being
used to provide access to the device the rootfs resides on.
When this module is installed into the ramfs it is triggered by
the udev event from the nic being setup (online).
Then if syslog is configured it is started and will forward all
kernel messages to the given syslog server.
The syslog implementation is detected automatically by finding the
apropriate binary with the following order:
rsyslogd
syslogd
syslog-ng
Then if detected the syslog.conf is generated and syslog is started.
Bootparameters:
syslogserver=ip Where to syslog to
sysloglevel=level What level has to be logged
syslogtype=rsyslog|syslog|syslogng
Don't auto detect syslog but set it
2009-08-19 10:40:02 +02:00
a607b7d4b8
Fix syntax error in conditional
...
Fix a typo in the conditionals. Not sure how I didn't notice these
before
2009-08-19 10:34:06 +02:00
1a4c4015bc
Modify nbd hostonly checking to use check_block_and_slaves.
...
This allows hostonly nbd checking to work in more exotic situations,
such as mdraid/dmraid/lvm/crypt on top of nbd.
2009-08-16 18:22:21 -05:00
d7df77d172
Modified lvm hostonly checking to use dracut-functions
2009-08-16 15:29:59 -05:00
7e2b4f8367
Make mdraid checking use common functionality in dracut-functions
2009-08-16 15:29:59 -05:00
bc9ae19deb
Make dmraid checking use common functionality in dracut-functions
2009-08-16 15:29:50 -05:00
17829e94aa
Move common rootfs finding and backing store tree checking to dracut-functions
2009-08-16 15:29:50 -05:00
4e207ec862
The network check script sources , so it must use /bin/bash
2009-08-16 15:29:49 -05:00
b5ccf21d2b
Make the lvm check script really check to see if root is on an lvm device
...
somehow when run in hostonly mode.
2009-08-16 15:29:42 -05:00
62bde8ed7e
make mdraid check script really check to see of the rootfs is on an
...
mdraid block device when run in hostonly mode.
2009-08-16 15:29:27 -05:00
dc670134f8
Make hostonly check for dmraid only pass if root is really on
...
a dmraid volume somehow.
2009-08-16 15:29:18 -05:00
39d6cb5df4
Make hostonly checking in 90crypt only pass if root is really on an
...
encrypted block device somewhere along the chain.
2009-08-16 15:28:56 -05:00
ddfd1d10a0
Split kernel module loading into smaller chunks.
...
This prepares to more tightly integrate dracut-gencmdline with the
rest of the dracut scripts
2009-08-16 15:28:45 -05:00
7addf9d430
rpmversion: dash does not want as I do
2009-08-14 17:47:07 +02:00
6d0480426d
rpmversion: fixed rpm query
2009-08-14 17:37:00 +02:00
1c6e9e5c0a
new module 10rpmversion
...
outputs rpm version of dracut
2009-08-14 17:27:50 +02:00
0b610a002c
removed 90dmsetup
2009-08-14 17:27:49 +02:00
0f1361ee1c
udev-rules: hardcode helper binaries
2009-08-14 17:27:49 +02:00
b806825d2a
nfs: only install libnfsidmap_nsswitch.so of all libnfsidmap_*.so
2009-08-14 17:27:49 +02:00
f905c1d7d6
dmsquash-live: install the "real" eject
...
eject might be a symlink to consolehelper
2009-08-14 17:27:49 +02:00
88838bf238
network: exclude kernel modules, which need iw_handler_get_spy
2009-08-14 16:21:03 +02:00
ed2de829e1
cryptroot-ask.sh: fix rd_LUKS_UUID handling
2009-08-14 16:21:03 +02:00
ff34c9dcd4
10redhat-i18n: remove unnecessary files
2009-08-14 16:21:03 +02:00
3d4641fd98
add rdinfo command line parameter
...
print informational output though "quiet" is set
2009-08-14 16:21:03 +02:00
ed6220fd25
plymouth: make the daemon attach to the session/terminal
...
https://bugzilla.redhat.com/show_bug.cgi?id=516104
it can be so simple :-)
2009-08-07 12:08:31 +02:00
148284ced6
Remove ugly hack
2009-08-06 11:07:43 -04:00
3714170f33
plymouth-populate-initrd: do not install label.so
...
https://bugzilla.redhat.com/show_bug.cgi?id=515589
It ends up installing the label.so control plugin which isn't supposed
to get installed into the initrd. this makes cairo and libX11 and all sorts of
things move into the initrd that aren't supposed to.
2009-08-06 14:43:13 +02:00
a15ad925e9
support for unusual terminfo location
...
the following patch makes the 95terminfo module work on gentoo.
2009-08-06 14:43:12 +02:00
13d5225fc8
Temporary hack to provide ifcfg files to OS in a way that might work with existing initscripts.
...
I am not happy about this. It shouldn't be the job of dracut to do this. The initscripts should
deal with the plain /dev/.initramfs/ifcfg/ directory accordingly. Doing this for now because
notting insists upon it. We need to clean this up after we network option passing working.
2009-08-04 17:15:39 -04:00
cc0e7a369b
Parse options earlier for live root
2009-08-04 16:25:43 +02:00
7f819d7041
Support resetting the live overlay
...
If you're using a persistent overlay, you might want to reset it
at boot time if it has become corrupted somehow. Support using
reset_overlay as a command line optino to do so
2009-08-04 16:25:43 +02:00
dfba82b3d3
And support passing in the persistent overlay
...
The persistent overlay can be specified with an overlay= argument
on the command line. We'll probably try to move this into the
root= syntax soon, but this is the old way that works
2009-08-04 16:25:43 +02:00
3ae2f09a04
Support passing live_ram
...
live_ram allows you to load the entirety of the live image into
RAM and run it from there rather than off of the block device
2009-08-04 16:25:43 +02:00
Harald Hoyer