fsck command is used to check and optionally repair one or more Linux file systems. You don’t have to create or alter/modify any file. All you need to do is just create empty file called /forcefsck.
How does fsck force work at every boot on a Linux?
On a Debian or Ubuntu Linux /etc/init.d/checkfs.sh (or /etc/init/mountall.conf on a Ubuntu LTS v14.04+) is used to run fsck command if /forcefsck.
Check the /etc/init/mountall.conf file, for example the file from Ubuntu 14.04 LTS
script
. /etc/default/rcS || true
[ -f /forcefsck ] && force_fsck="--force-fsck"
[ "$FSCKFIX" = "yes" ] && fsck_fix="--fsck-fix"
# Doesn not work so well if mountall is responsible for mounting /proc, heh.
if [ -e /proc/cmdline ]; then
read line < /proc/cmdline
for arg in $line; do
case $arg in
-q|--quiet|-v|--verbose|--debug)
debug_arg=$arg
;;
esac
done < /proc/cmdline
fi
# set $LANG so that messages appearing in plymouth are translated
if [ -r /etc/default/locale ]; then
. /etc/default/locale || true
export LANG LANGUAGE LC_MESSAGES LC_ALL
fi
exec mountall --daemon $force_fsck $fsck_fix $debug_arg |
On a RHEL/CentOS 6.x /etc/rc.sysinit file is used to run fsck command. If there is a /fsckoptions file, the options are loaded from the /fsckoptions file. If there is a /forcefsck file, -f is added.
Check the /etc/rc.sysinit file, for example the file from RHEL/CentOS Linux 6.x
Force fsck on boot using /forcefsck
By creating /forcefsck file you will force the Linux system (or rc scripts) to perform a full file system check. First, login as the root user:
$ su -
OR
$ sudo -s
Change directory to root (/) directory [optional]:
# cd /
Create a file called forcefsck:
# touch /forcefsck
Now, reboot the system:
# reboot
Frce fsck on next boot using shutdown command (may not work on many modern distros)
The -F option force fsck on reboot, login as root and type the following command to reboot and run fsck:
# shutdown -rF now
The above will check all the relevant partitions immediately.
A note about systemd based systems
The above mentioned solution only works with the old SysVinit and early versions of Upstart. It won’t work with systemd. systemd-fsck understands one kernel command line parameter:
fsck.mode=
One of “auto”, “force”, “skip”. Controls the mode of operation. The default is “auto”, and ensures that file system checks are done when the file system checker deems them necessary. “force” unconditionally results in full file system checks. “skip” skips any file system checks.fsck.repair=
One of “preen”, “yes”, “no”. Controls the mode of operation. The default is ” preen”, and will automatically repair problems that can be safely fixed. “yes ” will answer yes to all questions by fsck and “no” will answer no to all questions.
30 comment
Good info. You don’t need to goto the root directory if you’re providing the full path, as you are. You can:
touch /forcefsck
from anywhere… You do need to cd / if you are using a relative path, such as
cd /
touch forcefsck
or
cd /
touch ./forcefsk
but not
cd ~
touch ../forcefsk
(changes to /root, and creates the file one directory level higher)
nice! didn’t know that at alL!!
Hi every one,
I’m able to reboot my target board using the command [reboot] from the bash prompt.
But i can able to by the command [reboot -f].
What should i need to do if i want to use the command [reboot]
Plz help me.
Thanks in Advance
shutdown -rF now <-will reboot and forcefsck now. More info: man shutdown
@dj:
Your ‘shutdown’ tip won’t work on many modern distributions, since they use Upstart instead of the old Sysvinit system.
Upstart has dropped many legacy commands that would be better done differently, in favor of a modern lean system.
‘touch /forcefsck’ is currently the most universal way.
On RHEL 4, ‘shutdown -rF’ just writes /forcefsck anyway.
There’s no need to use the touch command, “>/forcefsck” will do.
thanks! sudo touch /forcefsck is a good way :)
What about trying to force fsck another partition ?
then u could just umount it……
umount /dev/hdb2; fsck -y -c -f /dev/hdb2;echo ‘w00t’
Thanks, this works great for me!
[email protected]:/$ sudo umount /dev/hdb2
umount: /home: device is busy.
[email protected]:/$ echo ‘unwoot’
Will performing a “shutdown -rF now” or a “touch /forcefsck ” at a specific date and time once a week prevent the “automatic fsck” that is forced by the ext2/ext3 partition parameters ?? (can be set by tune2fs ) – i mean.. will it “reset” the timeouts/deadlines if doing a force fsck manualy once a week so there will be no “surprise” when trying to reboot my server if needed ?? any positive experience on that?
Hey, thanks for the information. Appreciate it.
Running fsck, by whatever means, will cause the mount counter to be reset to 0. Well, so long as it completes successfully. If fsck dies for some reason (like trying to run on a mounted or read-only file system, oops) then the counter is not reset.
As for running on fsck on the partition mounted on /home, that should automagically mount, and will be checked at that time; just do (no quotes) “sudo touch /home/forcefsck” if mounted and if not mounted, you can run it directly, if you can log in, that is. Get to root and just run it one the device: “sudo fsck /dev/sda6” or whatever device it is.
Hi ,
I am going to check your tip hoping that ,this will work on My Ubuntu Karmic System with Ext4 File System.
And Yes ,it Worked with Ext4 FS.Thanks for the Tip!
I used “touch /forcefsck” which forced the system to check after reboot…fine
but on the subsequent boots still forces to check. I looked in root dir for the forcefsck file but it is not there….what then makes the system force to check? :(
How to specify some fsck cli params in this case?
as root, do the following
# (mount partition as read only – allows fsck to work)
mount -oremount,ro /
e2fsck /dev/sda1
where / is the path to device you want to fsck, and sda1 is the device name (could be hda1, hda2, sdb3 etc)
then when done, do
mount -oremount,rw /
telinit
(here it will give you a number which is current run level, note it (debian is N 2)
then.
telinit 1; telinit 2
(where 2 is the number given to you in first telinit line)
This b
Don’t forget to put a ‘y’ in the forcefsck file if you want to scan a remote drive. If not the check will be stuck while waiting for you to answer ‘y’, but you won’t be able to see that question…
so:
On Fedora 17 the -F option is not available for command shutdown.
can we do this on specific filesystem ?
Note : updated fstab fields as 0 2 ,still server skip fsck on …..
I followed the directions and it worked for my Mythbuntu system…on EVERY reboot.
I found I had to remove the /forcefsck file from my hard drive, when I wanted the disk checks at reboot to stop.
BUT, I could not see the /forcefsck file in the / directory until I did an “ls -a”
Best wishes!
You could also use tune2fs and set the filesystem mount count to something higher than the limit which triggers the filesystem check. This is very useful to check other filesystems besides the root /
# tune2fs -l /dev/sda1
and you can see on the output:
Mount count: 3
Maximum mount count: 38
so you have to place the Mount count higher than 38 with the following command:
# tune2fs -C 39 /dev/sda1
check the value changed with tune2fs -l and then reboot!
On my servers I like to set the filesystem Maximum mount count to 1, because when the server ever gets shutdown some shit did happen! So we must check everything!!
Use “sudo chattr +i /forcefsck” without the quotes if you want to perform a filesystem check every time the Raspberry PI boots, I do just to check for problems with the SD card. The +i makes the file forcefsck immutable so it doesn’t get removed even after a reboot. I found the file disappeared after every reboot but this fixed that problem.
Hope this stops someone else wondering where the file goes and how stop it happening.
Is there a way of setting this up to automatically repair any issues that it finds? Also, in the comment about adding a “y” to the file to scan remote drives, does that mean networked drives (located on another computer) or does it mean other drives in the same computer?
I have two internal hard drives, and four USB (External) drives attached to a server, and I want (need?) to fsck at least one or two of the external drives–but I want to do all of the drives.
In our environment some of the servers Filesystem automatically mount in readonly mode on the fly, we fixed it by running forefsck at reboot. may i know what will cause this kind of issue??
Check disk health using smartctl command.