9 dracut.cmdline - dracut kernel command line options
13 The root device used by the kernel is specified in the boot configuration
14 file on the kernel command line, as always.
16 The traditional _root=/dev/sda1_ style device specification is allowed, but not
17 encouraged. The root device should better be identified by LABEL or UUID. If a
18 label is used, as in _root=LABEL=<label_of_root>_ the initramfs will search all
19 available devices for a filesystem with the appropriate label, and mount that
20 device as the root filesystem. _root=UUID=<uuidnumber>_ will mount the partition
21 with that UUID as the root filesystem.
23 In the following all kernel command line parameters, which are processed by
24 dracut, are described.
26 "rd.*" parameters mentioned without "=" are boolean parameters. They can be
27 turned on/off by setting them to {0|1}. If the assignment with "=" is missing
28 "=1" is implied. For example _rd.info_ can be turned off with _rd.info=0_ or
29 turned on with _rd.info=1_ or _rd.info_. The last value in the kernel command
30 line is the value, which is honored.
34 **init=**__<path to real init>__::
35 specify the path to the init program to be started after the initramfs has
38 **root=**__<path to blockdevice>__::
39 specify the block device to use as the root filesystem.
45 root=/dev/disk/by-path/pci-0000:00:1f.1-scsi-0:0:1:0-part1
46 root=/dev/disk/by-label/Root
48 root=/dev/disk/by-uuid/3f5ad593-4546-4a94-a374-bcfb68aa11f7
49 root=UUID=3f5ad593-4546-4a94-a374-bcfb68aa11f7
50 root=PARTUUID=3f5ad593-4546-4a94-a374-bcfb68aa11f7
53 **rootfstype=**__<filesystem type>__:: "auto" if not specified.
61 **rootflags=**__<mount options>__::
62 specify additional mount options for the root filesystem. If not set,
63 _/etc/fstab_ of the real root will be parsed for special mount options and
67 force mounting _/_ and _/usr_ (if it is a separate device) read-only. If
68 none of ro and rw is present, both are mounted according to _/etc/fstab_.
71 force mounting _/_ and _/usr_ (if it is a separate device) read-write.
74 **rootfallback=**__<path to blockdevice>__::
75 specify the block device to use as the root filesystem, if the normal root
76 cannot be found. This can only be a simple block device with a simple file
77 system, for which the filesystem driver is either compiled in, or added
78 manually to the initramfs. This parameter can be specified multiple times.
80 **rd.auto** **rd.auto=1**::
81 enable autoassembly of special devices like cryptoLUKS, dmraid, mdraid or
82 lvm. Default is off as of dracut version >= 024.
85 removes all compiled in configuration of the host system the initramfs image
86 was built on. This helps booting, if any disk layout changed, especially in
87 combination with rd.auto or other parameters specifying the layout.
90 prompts the user for additional kernel command line parameters
93 do not honor special mount options for the root filesystem found in
94 _/etc/fstab_ of the real root.
96 **resume=**__<path to resume partition>__::
97 resume from a swap partition
102 resume=/dev/disk/by-path/pci-0000:00:1f.1-scsi-0:0:1:0-part1
103 resume=/dev/disk/by-uuid/3f5ad593-4546-4a94-a374-bcfb68aa11f7
104 resume=UUID=3f5ad593-4546-4a94-a374-bcfb68aa11f7
108 skip fsck for rootfs and _/usr_. If you're mounting _/usr_ read-only and
109 the init system performs fsck before remount, you might want to use this
110 option to avoid duplication.
115 Using iso-scan/filename with a Fedora/Red Hat/CentOS Live iso should just work
116 by copying the original kernel cmdline parameters.
121 menuentry 'Live Fedora 20' --class fedora --class gnu-linux --class gnu --class os {
122 set isolabel=Fedora-Live-LXDE-x86_64-20-1
123 set isofile="/boot/iso/Fedora-Live-LXDE-x86_64-20-1.iso"
124 loopback loop $isofile
125 linux (loop)/isolinux/vmlinuz0 boot=isolinux iso-scan/filename=$isofile root=live:LABEL=$isolabel ro rd.live.image quiet rhgb
126 initrd (loop)/isolinux/initrd0.img
132 **rd.emergency=**__[reboot|poweroff|halt]__::
133 specify, what action to execute in case of a critical failure. rd.shell=0 also
136 **rd.driver.blacklist=**__<drivername>__[,__<drivername>__,...]::
137 do not load kernel module <drivername>. This parameter can be specified
140 **rd.driver.pre=**__<drivername>__[,__<drivername>__,...]::
141 force loading kernel module <drivername>. This parameter can be specified
144 **rd.driver.post=**__<drivername>__[,__<drivername>__,...]::
145 force loading kernel module <drivername> after all automatic loading modules
146 have been loaded. This parameter can be specified multiple times.
148 **rd.retry=**__<seconds>__::
149 specify how long dracut should retry the initqueue to configure devices.
150 The default is 30 seconds. After 2/3 of the time, degraded raids are force
151 started. If you have hardware, which takes a very long time to announce its
152 drives, you might want to extend this value.
154 **rd.timeout=**__<seconds>__::
155 specify how long dracut should wait for devices to appear. The
156 default is '0', which means 'forever'. Note that this timeout
157 should be longer than rd.retry to allow for proper configuration.
160 accept self-signed certificates for ssl downloads.
162 **rd.ctty=**__<terminal device>__::
163 specify the controlling terminal for the console.
164 This is useful, if you have multiple "console=" arguments.
166 [[dracutkerneldebug]]
169 If you are dropped to an emergency shell, the file
170 _/run/initramfs/rdsosreport.txt_ is created, which can be saved to a (to be
171 mounted by hand) partition (usually /boot) or a USB stick. Additional debugging
172 info can be produced by adding **rd.debug** to the kernel command line.
173 _/run/initramfs/rdsosreport.txt_ contains all logs and the output of some tools.
174 It should be attached to any report about dracut problems.
177 print informational output though "quiet" is set
180 allow dropping to a shell, if root mounting fails
183 set -x for the dracut shell.
184 If systemd is active in the initramfs, all output is logged to the systemd
185 journal, which you can inspect with "journalctl -ab".
186 If systemd is not active, the logs are written to dmesg and
187 _/run/initramfs/init.log_.
188 If "quiet" is set, it also logs to the console.
190 **rd.memdebug=[0-4]**::
191 Print memory usage info at various points, set the verbose level from 0 to 4.
193 Higher level means more debugging output:
197 1 - partial /proc/meminfo
199 3 - /proc/meminfo + /proc/slabinfo
200 4 - /proc/meminfo + /proc/slabinfo + tracekomem
201 NOTE: tracekomem is a shell script utilizing kernel trace to track
202 the rough total memory consumption of kernel modules during
203 loading. It may override other trace configurations.
207 drop to a shell at the end
209 **rd.break=**__{cmdline|pre-udev|pre-trigger|initqueue|pre-mount|mount|pre-pivot|cleanup}__::
210 drop to a shell on defined breakpoint
213 set udev to loglevel info
216 set udev to loglevel debug
220 **rd.vconsole.keymap=**__<keymap base file name>__::
221 keyboard translation table loaded by loadkeys; taken from keymaps directory;
222 will be written as KEYMAP to _/etc/vconsole.conf_ in the initramfs.
227 rd.vconsole.keymap=de-latin1-nodeadkeys
230 **rd.vconsole.keymap.ext=**__<list of keymap base file names>__::
231 list of extra keymaps to bo loaded (sep. by space); will be written as
232 EXT_KEYMAP to _/etc/vconsole.conf_ in the initramfs
234 **rd.vconsole.unicode**::
235 boolean, indicating UTF-8 mode; will be written as UNICODE to
236 _/etc/vconsole.conf_ in the initramfs
238 **rd.vconsole.font=**__<font base file name>__::
239 console font; taken from consolefonts directory; will be written as FONT to
240 _/etc/vconsole.conf_ in the initramfs.
245 rd.vconsole.font=LatArCyrHeb-16
248 **rd.vconsole.font.map=**__<console map base file name>__::
249 see description of '-m' parameter in setfont manual; taken from consoletrans
250 directory; will be written as FONT_MAP to _/etc/vconsole.conf_ in the
253 **rd.vconsole.font.unimap=**__<unicode table base file name>__::
254 see description of '-u' parameter in setfont manual; taken from unimaps
255 directory; will be written as FONT_UNIMAP to _/etc/vconsole.conf_ in the
258 **rd.locale.LANG=**__<locale>__::
259 taken from the environment; if no UNICODE is defined we set its value in
260 basis of LANG value (whether it ends with ".utf8" (or similar) or not); will
261 be written as LANG to _/etc/locale.conf_ in the initramfs.
266 rd.locale.LANG=pl_PL.utf8
269 **rd.locale.LC_ALL=**__<locale>__::
270 taken from the environment; will be written as LC_ALL to _/etc/locale.conf_
276 disable LVM detection
278 **rd.lvm.vg=**__<volume group name>__::
279 only activate the volume groups with the given name. rd.lvm.vg can be
280 specified multiple times on the kernel command line.
282 **rd.lvm.lv=**__<logical volume name>__::
283 only activate the logical volumes with the given name. rd.lvm.lv can be
284 specified multiple times on the kernel command line.
287 remove any _/etc/lvm/lvm.conf_, which may exist in the initramfs
292 disable crypto LUKS detection
294 **rd.luks.uuid=**__<luks uuid>__::
295 only activate the LUKS partitions with the given UUID. Any "luks-" of the
296 LUKS UUID is removed before comparing to _<luks uuid>_.
297 The comparisons also matches, if _<luks uuid>_ is only the beginning of the
298 LUKS UUID, so you don't have to specify the full UUID.
299 This parameter can be specified multiple times.
301 **rd.luks.allow-discards=**__<luks uuid>__::
302 Allow using of discards (TRIM) requests for LUKS partitions with the given
303 UUID. Any "luks-" of the LUKS UUID is removed before comparing to
304 _<luks uuid>_. The comparisons also matches, if _<luks uuid>_ is only the
305 beginning of the LUKS UUID, so you don't have to specify the full UUID.
306 This parameter can be specified multiple times.
308 **rd.luks.allow-discards**::
309 Allow using of discards (TRIM) requests on all LUKS partitions.
311 **rd.luks.crypttab=0**::
312 do not check, if LUKS partition is in _/etc/crypttab_
314 **rd.luks.timeout=**__<seconds>__::
315 specify how long dracut should wait when waiting for the user to enter the
316 password. This avoid blocking the boot if no password is entered. It does
317 not apply to luks key. The default is '0', which means 'forever'.
319 crypto LUKS - key on removable device support
320 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
321 **rd.luks.key=**__<keypath>__:__<keydev>__:__<luksdev>__::
322 _keypath_ is a path to key file to look for. It's REQUIRED. When _keypath_
323 ends with '.gpg' it's considered to be key encrypted symmetrically with GPG.
324 You will be prompted for password on boot. GPG support comes with
325 'crypt-gpg' module which needs to be added explicitly.
327 _keydev_ is a device on which key file resides. It might be kernel name of
328 devices (should start with "/dev/"), UUID (prefixed with "UUID=") or label
329 (prefix with "LABEL="). You don't have to specify full UUID. Just its beginning
330 will suffice, even if its ambiguous. All matching devices will be probed.
331 This parameter is recommended, but not required. If not present, all block
332 devices will be probed, which may significantly increase boot time.
334 If _luksdev_ is given, the specified key will only be applied for that LUKS
335 device. Possible values are the same as for _keydev_. Unless you have several
336 LUKS devices, you don't have to specify this parameter. The simplest usage is:
341 rd.luks.key=/foo/bar.key
344 As you see, you can skip colons in such a case.
347 ===============================
348 dracut pipes key to cryptsetup with _-d -_ argument, therefore you need to pipe
349 to crypsetup luksFormat with _-d -_, too!
351 Here follows example for key encrypted with GPG:
355 gpg --quiet --decrypt rootkey.gpg | \
356 cryptsetup -d - -v --cipher serpent-cbc-essiv:sha256 \
357 --key-size 256 luksFormat /dev/sda3
360 If you use plain keys, just add path to _-d_ option:
364 cryptsetup -d rootkey.key -v --cipher serpent-cbc-essiv:sha256 \
365 --key-size 256 luksFormat /dev/sda3
367 ===============================
372 disable MD RAID detection
375 disable MD RAID for imsm/isw raids, use DM RAID instead
378 disable MD RAID for SNIA ddf raids, use DM RAID instead
381 ignore mdadm.conf included in initramfs
383 **rd.md.waitclean=1**::
384 wait for any resync, recovery, or reshape activity to finish before
387 **rd.md.uuid=**__<md raid uuid>__::
388 only activate the raid sets with the given UUID. This parameter can be
389 specified multiple times.
394 disable DM RAID detection
396 **rd.dm.uuid=**__<dm raid uuid>__::
397 only activate the raid sets with the given UUID. This parameter can be
398 specified multiple times.
403 disable multipath detection
410 **boot=**__<boot device>__::
411 specify the device, where /boot is located.
417 boot=/dev/disk/by-path/pci-0000:00:1f.1-scsi-0:0:1:0-part1
422 **rd.fips.skipkernel**::
423 skip checksum check of the kernel image. Useful, if the kernel image is not
424 in a separate boot partition.
430 =====================
431 It is recommended to either bind an interface to a MAC with the **ifname**
432 argument, or to use the systemd-udevd predictable network interface names.
434 Predictable network interface device names based on:
436 - firmware/bios-provided index numbers for on-board devices
437 - firmware-provided pci-express hotplug slot index number
438 - physical/geographical location of the hardware
439 - the interface's MAC address
442 http://www.freedesktop.org/wiki/Software/systemd/PredictableNetworkInterfaceNames
444 Two character prefixes based on the type of interface:
452 o<index>:: on-board device index number
453 s<slot>[f<function>][d<dev_id>]:: hotplug slot index number
455 [P<domain>]p<bus>s<slot>[f<function>][d<dev_id>]:: PCI geographical location
456 [P<domain>]p<bus>s<slot>[f<function>][u<port>][..][c<config>][i<interface>]:: USB port number chain
458 All multi-function PCI devices will carry the [f<function>] number in the
459 device name, including the function 0 device.
461 When using PCI geography, The PCI domain is only prepended when it is not 0.
463 For USB devices the full chain of port numbers of hubs is composed. If the
464 name gets longer than the maximum number of 15 characters, the name is not
466 The usual USB configuration == 1 and interface == 0 values are suppressed.
468 PCI ethernet card with firmware index "1"::
471 PCI ethernet card in hotplug slot with firmware index number::
474 PCI ethernet multi-function card with 2 ports::
481 USB built-in 3G modem::
486 =====================
488 **ip=**__{dhcp|on|any|dhcp6|auto6|either6}__::
489 dhcp|on|any::: get ip from dhcp server from all interfaces. If root=dhcp,
490 loop sequentially through all interfaces (eth0, eth1, ...) and use the first
491 with a valid DHCP root-path.
493 auto6::: IPv6 autoconfiguration
497 either6::: if auto6 fails, then dhcp6
499 **ip=**__<interface>__:__{dhcp|on|any|dhcp6|auto6}__[:[__<mtu>__][:__<macaddr>__]]::
500 This parameter can be specified multiple times.
502 =====================
503 dhcp|on|any|dhcp6::: get ip from dhcp server on a specific interface
504 auto6::: do IPv6 autoconfiguration
505 <macaddr>::: optionally **set** <macaddr> on the <interface>. This
506 cannot be used in conjunction with the **ifname** argument for the
508 =====================
510 **ip=**__<client-IP>__:[__<peer>__]:__<gateway-IP>__:__<netmask>__:__<client_hostname>__:__<interface>__:__{none|off|dhcp|on|any|dhcp6|auto6|ibft}__[:[__<mtu>__][:__<macaddr>__]]::
511 explicit network configuration. If you want do define a IPv6 address, put it
512 in brackets (e.g. [2001:DB8::1]). This parameter can be specified multiple
513 times. __<peer>__ is optional and is the address of the remote endpoint
514 for pointopoint interfaces and it may be followed by a slash and a decimal
515 number, encoding the network prefix length.
517 =====================
518 <macaddr>::: optionally **set** <macaddr> on the <interface>. This
519 cannot be used in conjunction with the **ifname** argument for the
521 =====================
523 **ip=**__<client-IP>__:[__<peer>__]:__<gateway-IP>__:__<netmask>__:__<client_hostname>__:__<interface>__:__{none|off|dhcp|on|any|dhcp6|auto6|ibft}__[:[__<dns1>__][:__<dns2>__]]::
524 explicit network configuration. If you want do define a IPv6 address, put it
525 in brackets (e.g. [2001:DB8::1]). This parameter can be specified multiple
526 times. __<peer>__ is optional and is the address of the remote endpoint
527 for pointopoint interfaces and it may be followed by a slash and a decimal
528 number, encoding the network prefix length.
530 **ifname=**__<interface>__:__<MAC>__::
531 Assign network device name <interface> (ie "bootnet") to the NIC with
534 WARNING: Do **not** use the default kernel naming scheme for the interface name,
535 as it can conflict with the kernel names. So, don't use "eth[0-9]+" for the
536 interface name. Better name it "bootnet" or "bluesocket".
538 **rd.route=**__<net>__/__<netmask>__:__<gateway>__[:__<interface>__]::
539 Add a static route with route options, which are separated by a colon.
540 IPv6 addresses have to be put in brackets.
545 rd.route=192.168.200.0/24:192.168.100.222:ens10
546 rd.route=192.168.200.0/24:192.168.100.222
547 rd.route=192.168.200.0/24::ens10
548 rd.route=[2001:DB8:3::/8]:[2001:DB8:2::1]:ens10
551 **bootdev=**__<interface>__::
552 specify network interface to use routing and netroot information from.
553 Required if multiple ip= lines are used.
555 **BOOTIF=**__<MAC>__::
556 specify network interface to use routing and netroot information from.
559 Disable BOOTIF parsing, which is provided by PXE
561 **nameserver=**__<IP>__ [**nameserver=**__<IP>__ ...]::
562 specify nameserver(s) to use
565 Disable DNS setting of DHCP parameters.
568 boolean, turn off biosdevname network interface renaming
571 boolean, bring up network even without netroot set
573 **vlan=**__<vlanname>__:__<phydevice>__::
574 Setup vlan device named <vlanname> on <phydeivce>.
575 We support the four styles of vlan names: VLAN_PLUS_VID (vlan0005),
576 VLAN_PLUS_VID_NO_PAD (vlan5), DEV_PLUS_VID (eth0.0005),
577 DEV_PLUS_VID_NO_PAD (eth0.5)
579 **bond=**__<bondname>__[:__<bondslaves>__:[:__<options>__[:<mtu>]]]::
580 Setup bonding device <bondname> on top of <bondslaves>.
581 <bondslaves> is a comma-separated list of physical (ethernet) interfaces.
582 <options> is a comma-separated list on bonding options (modinfo bonding for
583 details) in format compatible with initscripts. If <options> includes
584 multi-valued arp_ip_target option, then its values should be separated by
585 semicolon. if the mtu is specified, it will be set on the bond master.
586 Bond without parameters assumes
587 bond=bond0:eth0,eth1:mode=balance-rr
589 **team=**__<teammaster>__:__<teamslaves>__::
590 Setup team device <teammaster> on top of <teamslaves>.
591 <teamslaves> is a comma-separated list of physical (ethernet) interfaces.
593 **bridge=**__<bridgename>__:__<ethnames>__::
594 Setup bridge <bridgename> with <ethnames>. <ethnames> is a comma-separated
595 list of physical (ethernet) interfaces. Bridge without parameters assumes
600 **root=**\[_<server-ip>_:]__<root-dir>__[:__<nfs-options>__]::
601 mount nfs share from <server-ip>:/<root-dir>, if no server-ip is given, use
602 dhcp next_server. If server-ip is an IPv6 address it has to be put in
603 brackets, e.g. [2001:DB8::1]. NFS options can be appended with the prefix
604 ":" or "," and are separated by ",".
606 **root=**nfs:\[_<server-ip>_:]__<root-dir>__[:__<nfs-options>__], **root=**nfs4:\[_<server-ip>_:]__<root-dir>__[:__<nfs-options>__], **root=**__{dhcp|dhcp6}__::
607 root=dhcp alone directs initrd to look at the DHCP root-path where NFS
608 options can be specified.
613 root-path=<server-ip>:<root-dir>[,<nfs-options>]
614 root-path=nfs:<server-ip>:<root-dir>[,<nfs-options>]
615 root-path=nfs4:<server-ip>:<root-dir>[,<nfs-options>]
618 **root=**_/dev/nfs_ nfsroot=\[_<server-ip>_:]__<root-dir>__[:__<nfs-options>__]::
619 _Deprecated!_ kernel Documentation_/filesystems/nfsroot.txt_ defines this
620 method. This is supported by dracut, but not recommended.
622 **rd.nfs.domain=**__<NFSv4 domain name>__::
623 Set the NFSv4 domain name. Will override the settings in _/etc/idmap.conf_.
625 **rd.net.dhcp.retry=**__<cnt>__::
626 If this option is set, dracut will try to connect via dhcp <cnt> times before failing.
629 **rd.net.timeout.dhcp=**__<arg>__::
630 If this option is set, dhclient is called with "-timeout <arg>".
632 **rd.net.timeout.iflink=**__<seconds>__::
633 Wait <seconds> until link shows up. Default is 60 seconds.
635 **rd.net.timeout.ifup=**__<seconds>__::
636 Wait <seconds> until link has state "UP". Default is 20 seconds.
638 **rd.net.timeout.route=**__<seconds>__::
639 Wait <seconds> until route shows up. Default is 20 seconds.
641 **rd.net.timeout.ipv6dad=**__<seconds>__::
642 Wait <seconds> until IPv6 DAD is finished. Default is 50 seconds.
644 **rd.net.timeout.ipv6auto=**__<seconds>__::
645 Wait <seconds> until IPv6 automatic addresses are assigned. Default is 40 seconds.
647 **rd.net.timeout.carrier=**__<seconds>__::
648 Wait <seconds> until carrier is recognized. Default is 5 seconds.
652 **root=**cifs://[__<username>__[:__<password>__]@]__<server-ip>__:__<root-dir>__::
653 mount cifs share from <server-ip>:/<root-dir>, if no server-ip is given, use
654 dhcp next_server. if server-ip is an IPv6 address it has to be put in
655 brackets, e.g. [2001:DB8::1]. If a username or password are not specified
656 as part of the root, then they must be passed on the command line through
659 WARNING: Passwords specified on the kernel command line are visible for all
660 users via the file _/proc/cmdline_ and via dmesg or can be sniffed on the
661 network, when using DHCP with DHCP root-path.
663 **cifsuser**=__<username>__::
664 Set the cifs username, if not specified as part of the root.
666 **cifspass**=__<password>__::
667 Set the cifs password, if not specified as part of the root.
669 WARNING: Passwords specified on the kernel command line are visible for all
670 users via the file _/proc/cmdline_ and via dmesg or can be sniffed on the
671 network, when using DHCP with DHCP root-path.
675 **root=**iscsi:[__<username>__:__<password>__[:__<reverse>__:__<password>__]@][__<servername>__]:[__<protocol>__]:[__<port>__][:[__<iscsi_iface_name>__]:[__<netdev_name>__]]:[__<LUN>__]:__<targetname>__::
676 protocol defaults to "6", LUN defaults to "0". If the "servername" field is
677 provided by BOOTP or DHCP, then that field is used in conjunction with other
678 associated fields to contact the boot server in the Boot stage. However, if
679 the "servername" field is not provided, then the "targetname" field is then
680 used in the Discovery Service stage in conjunction with other associated
682 link:$$http://tools.ietf.org/html/rfc4173#section-5$$[rfc4173].
684 WARNING: Passwords specified on the kernel command line are visible for all
685 users via the file _/proc/cmdline_ and via dmesg or can be sniffed on the
686 network, when using DHCP with DHCP root-path.
691 root=iscsi:192.168.50.1::::iqn.2009-06.dracut:target0
694 If servername is an IPv6 address, it has to be put in brackets:
699 root=iscsi:[2001:DB8::1]::::iqn.2009-06.dracut:target0
702 **root=**__???__ **netroot=**iscsi:[__<username>__:__<password>__[:__<reverse>__:__<password>__]@][__<servername>__]:[__<protocol>__]:[__<port>__][:[__<iscsi_iface_name>__]:[__<netdev_name>__]]:[__<LUN>__]:__<targetname>__ ...::
703 multiple netroot options allow setting up multiple iscsi disks:
709 netroot=iscsi:192.168.50.1::::iqn.2009-06.dracut:target0
710 netroot=iscsi:192.168.50.1::::iqn.2009-06.dracut:target1
713 If servername is an IPv6 address, it has to be put in brackets:
718 netroot=iscsi:[2001:DB8::1]::::iqn.2009-06.dracut:target0
721 WARNING: Passwords specified on the kernel command line are visible for all
722 users via the file _/proc/cmdline_ and via dmesg or can be sniffed on the
723 network, when using DHCP with DHCP root-path.
724 You may want to use rd.iscsi.firmware.
726 **root=**__???__ **rd.iscsi.initiator=**__<initiator>__ **rd.iscsi.target.name=**__<target name>__ **rd.iscsi.target.ip=**__<target ip>__ **rd.iscsi.target.port=**__<target port>__ **rd.iscsi.target.group=**__<target group>__ **rd.iscsi.username=**__<username>__ **rd.iscsi.password=**__<password>__ **rd.iscsi.in.username=**__<in username>__ **rd.iscsi.in.password=**__<in password>__::
727 manually specify all iscsistart parameter (see **+iscsistart --help+**)
729 WARNING: Passwords specified on the kernel command line are visible for all
730 users via the file _/proc/cmdline_ and via dmesg or can be sniffed on the
731 network, when using DHCP with DHCP root-path.
732 You may want to use rd.iscsi.firmware.
734 **root=**_???_ **netroot=**iscsi **rd.iscsi.firmware=1**::
735 will read the iscsi parameter from the BIOS firmware
737 **rd.iscsi.login_retry_max=**__<num>__::
738 maximum number of login retries
740 **rd.iscsi.param=**__<param>__::
741 <param> will be passed as "--param <param>" to iscsistart.
742 This parameter can be specified multiple times.
747 "netroot=iscsi rd.iscsi.firmware=1 rd.iscsi.param=node.session.timeo.replacement_timeout=30"
754 iscsistart -b --param node.session.timeo.replacement_timeout=30
757 **rd.iscsi.ibft** **rd.iscsi.ibft=1**:
758 Turn on iBFT autoconfiguration for the interfaces
760 **rd.iscsi.waitnet=0**:
761 Turn off waiting for all interfaces to be up before trying to login to the iSCSI targets.
763 **rd.iscsi.testroute=0**:
764 Turn off checking, if the route to the iSCSI target IP is possible before trying to login.
768 **fcoe=**__<edd|interface|MAC>__:__{dcb|nodcb}__:__{fabric|vn2vn}__::
769 Try to connect to a FCoE SAN through the NIC specified by _<interface>_ or
770 _<MAC>_ or EDD settings. The second argument specifies if DCB
771 should be used. The optional third argument specifies whether
772 fabric or VN2VN mode should be used.
773 This parameter can be specified multiple times.
775 NOTE: letters in the MAC-address must be lowercase!
779 **root=**??? **netroot=**nbd:__<server>__:__<port/exportname>__[:__<fstype>__[:__<mountopts>__[:__<nbdopts>__]]]::
780 mount nbd share from <server>.
783 If "exportname" instead of "port" is given the standard port is used.
784 Newer versions of nbd are only supported with "exportname".
786 **root=dhcp** with **dhcp** **root-path=**nbd:__<server>__:__<port/exportname>__[:__<fstype>__[:__<mountopts>__[:__<nbdopts>__]]]::
787 root=dhcp alone directs initrd to look at the DHCP root-path where NBD
788 options can be specified. This syntax is only usable in cases where you are
789 directly mounting the volume as the rootfs.
792 If "exportname" instead of "port" is given the standard port is used.
793 Newer versions of nbd are only supported with "exportname".
798 same syntax as the kernel module parameter (s390 only)
802 **rd.zfcp=**__<zfcp adaptor device bus ID>__,__<WWPN>__,__<FCPLUN>__::
803 rd.zfcp can be specified multiple times on the kernel command
806 **rd.zfcp=**__<zfcp adaptor device bus ID>__::
807 If NPIV is enabled and the 'allow_lun_scan' parameter to the zfcp
808 module is set to 'Y' then the zfcp adaptor will be initiating a
809 scan internally and the <WWPN> and <FCPLUN> parameters can be omitted.
814 rd.zfcp=0.0.4000,0x5005076300C213e9,0x5022000000000000
819 ignore zfcp.conf included in the initramfs
823 **rd.znet=**__<nettype>__,__<subchannels>__,__<options>__::
824 The whole parameter is appended to /etc/ccw.conf, which is used on
825 RHEL/Fedora with ccw_init, which is called from udev for certain
827 rd.znet can be specified multiple times on the kernel command line.
832 rd.znet=qeth,0.0.0600,0.0.0601,0.0.0602,layer2=1,portname=foo
833 rd.znet=ctc,0.0.0600,0.0.0601,protocol=bar
838 Dracut offers multiple options for live booted images:
840 =====================
841 SquashFS with read-only filesystem image::: The system will boot with a
842 read-only filesystem from the SquashFS and apply a writable Device-mapper
843 snapshot or an OverlayFS overlay mount for the read-only base filesystem. This
844 method ensures a relatively fast boot and lower RAM usage. Users **must be
845 careful** to avoid writing too many blocks to a snapshot volume. Once the
846 blocks of the snapshot overlay are exhausted, the root filesystem becomes
847 read-only and may cause application failures. The snapshot overlay file is
848 marked 'Overflow', and a difficult recovery is required to repair and enlarge
849 the overlay offline. Non-persistent overlays are sparse files in RAM that only
850 consume content space as required blocks are allocated. They default to an
851 apparent size of 32 GiB in RAM. The size can be adjusted with the
852 **rd.live.overlay.size=** kernel command line option.
854 The filesystem structure is traditionally expected to be:
858 squashfs.img | SquashFS from LiveCD .iso
861 |- rootfs.img | Filesystem image to mount read-only
863 /bin | Live filesystem
869 For OverlayFS mount overlays, the filesystem structure may also be a direct
870 compression of the root filesystem:
874 squashfs.img | SquashFS from LiveCD .iso
876 /bin | Live filesystem
882 Dracut uses one of the overlay methods of live booting by default. No
883 additional command line options are required other than **root=live:<URL>** to
884 specify the location of your squashed filesystem.
886 - The compressed SquashFS image can be copied during boot to RAM at
887 `/run/initramfs/squashed.img` by using the **rd.live.ram=1** option.
888 - A device with a persistent overlay can be booted read-only by using the
889 **rd.live.overlay.readonly** option on the kernel command line. This will
890 either cause a temporary, writable overlay to be stacked over a read-only
891 snapshot of the root filesystem or the OverlayFS mount will use an additional
892 lower layer with the root filesystem.
894 Uncompressed live filesystem image:::
895 When the live system was installed with the '--skipcompress' option of the
896 __livecd-iso-to-disk__ installation script for Live USB devices, the root
897 filesystem image, __rootfs.img__, is expanded on installation and no SquashFS
898 is involved during boot.
900 - If **rd.live.ram=1** is used in this situation, the full, uncompressed
901 root filesystem is copied during boot to `/run/initramfs/rootfs.img` in the
904 - If **rd.live.overlay=none** is provided as a kernel command line option,
905 a writable, linear Device-mapper target is created on boot with no overlay.
907 Writable filesystem image:::
908 The system will retrieve a compressed filesystem image, extract it to
909 `/run/initramfs/fsimg/rootfs.img`, connect it to a loop device, create a
910 writable, linear Device-mapper target at `/dev/mapper/live-rw`, and mount that
911 as a writable volume at `/`. More RAM is required during boot but the live
912 filesystem is easier to manage if it becomes full. Users can make a filesystem
913 image of any size and that size will be maintained when the system boots. There
914 is no persistence of root filesystem changes between boots with this option.
916 The filesystem structure is expected to be:
920 rootfs.tgz | Compressed tarball containing filesystem image
922 /rootfs.img | Filesystem image at /run/initramfs/fsimg/
924 /bin | Live filesystem
930 To use this boot option, ensure that **rd.writable.fsimg=1** is in your kernel
931 command line and add the **root=live:<URL>** to specify the location
932 of your compressed filesystem image tarball or SquashFS image.
933 =====================
935 **rd.writable.fsimg=**1::
936 Enables writable filesystem support. The system will boot with a fully
937 writable (but non-persistent) filesystem without snapshots __(see notes above
938 about available live boot options)__. You can use the **rootflags** option to
939 set mount options for the live filesystem as well __(see documentation about
940 rootflags in the **Standard** section above)__.
941 This implies that the whole image is copied to RAM before the boot continues.
943 NOTE: There must be enough free RAM available to hold the complete image.
945 This method is very suitable for diskless boots.
947 **root=**live:__<url>__::
948 Boots a live image retrieved from __<url>__. Requires the dracut 'livenet'
949 module. Valid handlers: __http, https, ftp, torrent, tftp__.
954 root=live:http://example.com/liveboot.img
955 root=live:ftp://ftp.example.com/liveboot.img
956 root=live:torrent://example.com/liveboot.img.torrent
959 **rd.live.debug=**1::
960 Enables debug output from the live boot process.
962 **rd.live.dir=**__<path>__::
963 Specifies the directory within the boot device where the squashfs.img or
964 rootfs.img can be found. By default, this is `/LiveOS`.
966 **rd.live.squashimg=**__<filename of SquashFS image>__::
967 Specifies the filename for a SquashFS image of the root filesystem.
968 By default, this is __squashfs.img__.
971 Copy the complete image to RAM and use this for booting. This is useful
972 when the image resides on, e.g., a DVD which needs to be ejected later on.
974 **rd.live.overlay={**__<devspec>__[:__{<pathspec>|auto}__]|__none__}::
975 Manage the usage of a permanent overlay.
978 * _<devspec>_ specifies the path to a device with a mountable filesystem.
979 * _<pathspec>_ is the path to a file within that filesystem, which shall be
980 used to persist the changes made to the device specified by the
981 **root=live:__<url>__** option.
983 The default _pathspec_, when _auto_ or no _:<pathspec>_ is given, is
984 `/<+++<b>rd.live.dir</b>+++>/overlay-<label>-<uuid>`, where _<label>_ is the
985 device LABEL, and _<uuid>_ is the device UUID.
986 * _none_ (the word itself) specifies that no overlay will be used, such as when
987 an uncompressed, writable live root filesystem is available.
989 If a persistent overlay __is detected__ at the standard LiveOS path, the
990 overlay & overlay type detected, whether Device-mapper or OverlayFS, will be
997 rd.live.overlay=/dev/sdb1:persistent-overlay.img
998 rd.live.overlay=UUID=99440c1f-8daa-41bf-b965-b7240a8996f4
1001 **rd.live.overlay.size=**__<size_MiB>__::
1002 Specifies a non-persistent Device-mapper overlay size in MiB. The default is
1005 **rd.live.overlay.readonly=**1::
1006 This is used to boot with a normally read-write persistent overlay in a
1007 read-only mode. With this option, either an additional, non-persistent,
1008 writable snapshot overlay will be stacked over a read-only snapshot,
1009 `/dev/mapper/live‑ro`, of the base filesystem with the persistent overlay, or a
1010 read-only loop device, in the case of a writable __rootfs.img__, or an OverlayFS
1011 mount will use the persistent overlay directory linked at `/run/overlayfs‑r` as
1012 an additional lower layer along with the base root filesystem and apply a
1013 transient, writable upper directory overlay, in order to complete the booted
1016 **rd.live.overlay.reset=**1::
1017 Specifies that a persistent overlay should be reset on boot. All previous root
1018 filesystem changes are vacated by this action.
1020 **rd.live.overlay.thin=**1::
1021 Enables the usage of thin snapshots instead of classic dm snapshots.
1022 The advantage of thin snapshots is that they support discards, and will free
1023 blocks that are not claimed by the filesystem. In this use case, this means
1024 that memory is given back to the kernel when the filesystem does not claim it
1027 **rd.live.overlay.overlayfs=**1::
1028 Enables the use of the *OverlayFS* kernel module, if available, to provide a
1029 copy-on-write union directory for the root filesystem. OverlayFS overlays are
1030 directories of the files that have changed on the read-only base (lower)
1031 filesystem. The root filesystem is provided through a special overlay type
1032 mount that merges the lower and upper directories. If an OverlayFS upper
1033 directory is not present on the boot device, a tmpfs directory will be created
1034 at `/run/overlayfs` to provide temporary storage. Persistent storage can be
1035 provided on vfat or msdos formatted devices by supplying the OverlayFS upper
1036 directory within an embedded filesystem that supports the creation of trusted.*
1037 extended attributes and provides a valid d_type in readdir responses, such as
1038 with ext4 and xfs. On non-vfat-formatted devices, a persistent OverlayFS
1039 overlay can extend the available root filesystem storage up to the capacity of
1040 the LiveOS disk device.
1042 If a persistent overlay is detected at the standard LiveOS path, the overlay &
1043 overlay type detected, whether OverlayFS or Device-mapper, will be used.
1045 The **rd.live.overlay.readonly** option, which allows a persistent overlayfs to
1046 be mounted read-only through a higher level transient overlay directory, has
1047 been implemented through the multiple lower layers feature of OverlayFS.
1052 **rd.zipl=**__<path to blockdevice>__::
1053 Update the dracut commandline with the values found in the
1054 _dracut-cmdline.conf_ file on the given device.
1055 The values are merged into the existing commandline values
1056 and the udev events are regenerated.
1061 rd.zipl=UUID=0fb28157-99e3-4395-adef-da3f7d44835a
1066 **rd.cio_accept=**__<device-ids>__::
1067 Remove the devices listed in <device-ids> from the default
1068 cio_ignore kernel command-line settings.
1069 <device-ids> is a list of comma-separated CCW device ids.
1070 The default for this value is taken from the
1071 _/boot/zipl/active_devices.txt_ file.
1076 rd.cio_accept=0.0.0180,0.0.0800,0.0.0801,0.0.0802
1079 Plymouth Boot Splash
1080 ~~~~~~~~~~~~~~~~~~~~
1081 **plymouth.enable=0**::
1082 disable the plymouth bootsplash completely.
1085 disable the plymouth bootsplash only for the initramfs.
1089 **masterkey=**__<kernel master key path name>__::
1090 Set the path name of the kernel master key.
1095 masterkey=/etc/keys/kmk-trusted.blob
1098 **masterkeytype=**__<kernel master key type>__::
1099 Set the type of the kernel master key.
1104 masterkeytype=trusted
1107 **evmkey=**__<EVM key path name>__::
1108 Set the path name of the EVM key.
1113 evmkey=/etc/keys/evm-trusted.blob
1116 **ecryptfskey=**__<eCryptfs key path name>__::
1117 Set the path name of the eCryptfs key.
1122 ecryptfskey=/etc/keys/ecryptfs-trusted.blob
1125 Deprecated, renamed Options
1126 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
1127 Here is a list of options, which were used in dracut prior to version 008, and
1128 their new replacement.
1135 rd_DASD_MOD:: rd.dasd
1139 rdinitdebug rdnetdebug:: rd.debug
1143 rd_DM_UUID:: rd.dm.uuid
1145 rdblacklist:: rd.driver.blacklist
1147 rdinsmodpost:: rd.driver.post
1149 rdloaddriver:: rd.driver.pre
1151 rd_NO_FSTAB:: rd.fstab=0
1155 check:: rd.live.check
1157 rdlivedebug:: rd.live.debug
1159 live_dir:: rd.live.dir
1161 liveimg:: rd.live.image
1163 overlay:: rd.live.overlay
1165 readonly_overlay:: rd.live.overlay.readonly
1167 reset_overlay:: rd.live.overlay.reset
1169 live_ram:: rd.live.ram
1171 rd_NO_CRYPTTAB:: rd.luks.crypttab=0
1173 rd_LUKS_KEYDEV_UUID:: rd.luks.keydev.uuid
1175 rd_LUKS_KEYPATH:: rd.luks.keypath
1177 rd_NO_LUKS:: rd.luks=0
1179 rd_LUKS_UUID:: rd.luks.uuid
1181 rd_NO_LVMCONF:: rd.lvm.conf
1183 rd_LVM_LV:: rd.lvm.lv
1185 rd_NO_LVM:: rd.lvm=0
1187 rd_LVM_SNAPSHOT:: rd.lvm.snapshot
1189 rd_LVM_SNAPSIZE:: rd.lvm.snapsize
1191 rd_LVM_VG:: rd.lvm.vg
1193 rd_NO_MDADMCONF:: rd.md.conf=0
1195 rd_NO_MDIMSM:: rd.md.imsm=0
1199 rd_MD_UUID:: rd.md.uuid
1201 rd_NO_MULTIPATH: rd.multipath=0
1203 rd_NFS_DOMAIN:: rd.nfs.domain
1205 iscsi_initiator:: rd.iscsi.initiator
1207 iscsi_target_name:: rd.iscsi.target.name
1209 iscsi_target_ip:: rd.iscsi.target.ip
1211 iscsi_target_port:: rd.iscsi.target.port
1213 iscsi_target_group:: rd.iscsi.target.group
1215 iscsi_username:: rd.iscsi.username
1217 iscsi_password:: rd.iscsi.password
1219 iscsi_in_username:: rd.iscsi.in.username
1221 iscsi_in_password:: rd.iscsi.in.password
1223 iscsi_firmware:: rd.iscsi.firmware=0
1225 rd_NO_PLYMOUTH:: rd.plymouth=0
1231 rd_NO_SPLASH:: rd.splash
1233 rdudevdebug:: rd.udev.debug
1235 rdudevinfo:: rd.udev.info
1237 rd_NO_ZFCPCONF:: rd.zfcp.conf=0
1243 KEYMAP:: vconsole.keymap
1245 KEYTABLE:: vconsole.keymap
1247 SYSFONT:: vconsole.font
1249 CONTRANS:: vconsole.font.map
1251 UNIMAP:: vconsole.font.unimap
1253 UNICODE:: vconsole.unicode
1255 EXT_KEYMAP:: vconsole.keymap.ext
1257 Configuration in the Initramfs
1258 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1260 Any files found in _/etc/conf.d/_ will be sourced in the initramfs to
1261 set initial values. Command line options will override these values
1262 set in the configuration files.
1265 Can contain additional command line options. Deprecated, better use
1266 /etc/cmdline.d/*.conf.
1268 _/etc/cmdline.d/*.conf_::
1269 Can contain additional command line options.
1277 *dracut*(8) *dracut.conf*(5)