mkfs utility is used to create filesystem (ext2, ext3, ext4, etc) on your Linux system. You should specify the device name to mkfs on which the filesystem to be created.
WARNING: Executing these commands will destroy all the data on your filesystem. So, try these commands only on a test system where you don’t care about losing your data.
1. View the Available Filesystem Builder Commands
The filesystem builders (mkfs* commands) are usually searched in directories like /sbin/, /sbin/fs, /sbin/fs.d, /etc/fs and /etc. If not found, finally it searches the directories found in the PATH variable.
The following list shows the available mkfs* commands in a system.
# cd /sbin # ls mkfs* mkfs mkfs.bfs mkfs.cramfs mkfs.ext2 mkfs.ext3 mkfs.ext4 mkfs.ext4dev mkfs.minix mkfs.msdos mkfs.ntfs mkfs.vfat
If you are new to filesystem, read about Ext2 vs Ext3 vs Ext4.
2. Build a Filesystem on a Specific Device
In order to build the filesystem using mkfs command, the required arguments are device-filename and filesystem-type as shown below. The following example creates ext3 filesystem on /dev/sda6 partition.
# mkfs -t ext3 /dev/sda6 mke2fs 1.42 (29-Nov-2011) Filesystem label= OS type: Linux Block size=4096 (log=2) Fragment size=4096 (log=2) Stride=0 blocks, Stripe width=0 blocks 1120112 inodes, 4476416 blocks 223820 blocks (5.00%) reserved for the super user First data block=0 Maximum filesystem blocks=0 137 block groups 32768 blocks per group, 32768 fragments per group 8176 inodes per group Superblock backups stored on blocks: 32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, 2654208, 4096000 Allocating group tables: done Writing inode tables: done Creating journal (32768 blocks): done Writing superblocks and filesystem accounting information: done
Please note that the default filesystem type for mkfs command is ext2. If you don’t specify “-t” option, it will create ext2 filesystem.
Also, you can use the method we discussed earlier to identify whether you have ext2 or ext3 or ext4 file system.
3. Create a Filesystem with Journal using -j option
The example given below creates a file system with journaling.
# mke2fs /dev/sda6 -j mke2fs 1.42 (29-Nov-2011) Filesystem label= OS type: Linux Block size=4096 (log=2) Fragment size=4096 (log=2) Stride=0 blocks, Stripe width=0 blocks 1120112 inodes, 4476416 blocks 223820 blocks (5.00%) reserved for the super user First data block=0 Maximum filesystem blocks=0 137 block groups 32768 blocks per group, 32768 fragments per group 8176 inodes per group Superblock backups stored on blocks: 32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, 2654208, 4096000 Allocating group tables: done Writing inode tables: done Creating journal (32768 blocks): done Writing superblocks and filesystem accounting information: done
When you don’t need the journaling on a filesystem, then execute mke2fs command as shown below.
# mke2fs /dev/sda6
4. Check for Bad Blocks on a Device
Before creating a filesystem, you can check to make sure there are no bad blocks on the system using -c option as shown below. This will do a read-only test on the filesystem to make sure there are no bad blocks. You can also perform a read-write test to verify there are no bad blocks using “-c -c” option. Please note that this will be slower than the “-c” option.
# mke2fs -c /dev/sda6 mke2fs 1.42 (29-Nov-2011) Filesystem label= OS type: Linux Block size=4096 (log=2) Fragment size=4096 (log=2) Stride=0 blocks, Stripe width=0 blocks 1120112 inodes, 4476416 blocks 223820 blocks (5.00%) reserved for the super user First data block=0 Maximum filesystem blocks=0 137 block groups 32768 blocks per group, 32768 fragments per group 8176 inodes per group Superblock backups stored on blocks: 32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, 2654208, 4096000
5. Force to Create a Filesystem on a Mounted Partition
It is recommended not to do any filesystem operations on a mounted partition. But still you can force it to create a filesystem on in-use partition or mounted partition with option -F as shown below.
# mke2fs -F /dev/sda6
6. Bytes per Inode Group on a Partition
You may want to list out the used and available inodes in the partition as shown below using df command.
# df -i /dev/sda6 Filesystem Inodes IUsed IFree IUse% Mounted on /dev/sda6 1120112 11 1120101 1% /mydata
As seen above, on /dev/sda6 the IUsed is 11 out of 1120112. After the new file gets created on it, the value of IUsed gets changes accordingly as shown below.
# cd /mydata # touch sample.txt # ls sample.txt # df -i /dev/sda6 Filesystem Inodes IUsed IFree IUse% Mounted on /dev/sda6 1120112 12 1120100 1% /mydata
You may also use the tune2fs command to view the Inode details:
# tune2fs -l /dev/sda6 | grep Inode Inode count: 1120112 Inodes per group: 8176 Inode blocks per group: 511 Inode size: 256
To change the inode-per-group of the above partition(i.e:/dev/sda6) use -i option as follows:
# mke2fs -i 8192 /dev/sda6 mke2fs 1.42 (29-Nov-2011) Filesystem label= OS type: Linux Block size=4096 (log=2) Fragment size=4096 (log=2) Stride=0 blocks, Stripe width=0 blocks 2240224 inodes, 4476416 blocks 223820 blocks (5.00%) reserved for the super user First data block=0 Maximum filesystem blocks=0 137 block groups 32768 blocks per group, 32768 fragments per group 16352 inodes per group Superblock backups stored on blocks: 32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, 2654208, 4096000 Allocating group tables: done Writing inode tables: done Writing superblocks and filesystem accounting information: done
As seen above, the bytes-per-inode is changed to 8192 from the default value of 16384. Now the inode limit of the partition got increased as shown below.
# df -i /dev/sda6 Filesystem Inodes IUsed IFree IUse% Mounted on /dev/sda6 2240224 11 2240213 1% /mydata # tune2fs -l /dev/sda6 | grep Inode Inode count: 2240224 Inodes per group: 16352 Inode blocks per group: 1022 Inode size: 256
Changing the value of inodes-per-group just nothing but impacting the number of files to be kept on the partition. Reducing this value is for increasing the number of files on a partition.
7. Change the Inode size on a Partition
It is possible to change the size of an Inode on a parition using the option I and i as shown below.
# mke2fs -I 128 -i 8192 /dev/sda6 mke2fs 1.42 (29-Nov-2011) Filesystem label= OS type: Linux Block size=4096 (log=2) Fragment size=4096 (log=2) Stride=0 blocks, Stripe width=0 blocks 2240224 inodes, 4476416 blocks 223820 blocks (5.00%) reserved for the super user First data block=0 Maximum filesystem blocks=0 137 block groups 32768 blocks per group, 32768 fragments per group 16352 inodes per group Superblock backups stored on blocks: 32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, 2654208, 4096000 Allocating group tables: done Writing inode tables: done Writing superblocks and filesystem accounting information: done
As seen above, the size is being changed from 256 to 128 and it leads in increasing the total Inode count on a partition:
# tune2fs -l /dev//sda6 | grep Inode Inode count: 2240224 Inodes per group: 16352 Inode blocks per group: 511 Inode size: 128
8. Set the Volume Label for Partition
You can create a name/label for a partition using option -L. In the following example, we are assigning DATA as the label for /dev/sda6 partition.
# mke2fs -L DATA /dev/sda6 mke2fs 1.42 (29-Nov-2011) Filesystem label=DATA OS type: Linux Block size=4096 (log=2) Fragment size=4096 (log=2) Stride=0 blocks, Stripe width=0 blocks 1120112 inodes, 4476416 blocks 223820 blocks (5.00%) reserved for the super user First data block=0 Maximum filesystem blocks=0 137 block groups 32768 blocks per group, 32768 fragments per group 8176 inodes per group Superblock backups stored on blocks: 32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, 2654208, 4096000 Allocating group tables: done Writing inode tables: done Writing superblocks and filesystem accounting information: done
You can view the label name of the above partition by using e2label command as shown below:
# e2label /dev/sda6 DATA
Another way to view the label of a partition is using blkid command as shown below:
# blkid /dev/sda6 /dev/sda6: LABEL="DATA" UUID="0de74d35-6050-4838-99b0-46cb1d518da8" TYPE="ext2"
9. Simulate a Filesystem Creation
When executing mkfs command, it displays the contents like what it would do on stdout. You can even display the those messages on stdout without actually creating a filesystem using -n option as shown below:
# mkfs -t ext3 -n /dev/sda6 mke2fs 1.42 (29-Nov-2011) Filesystem label= OS type: Linux Block size=4096 (log=2) Fragment size=4096 (log=2) Stride=0 blocks, Stripe width=0 blocks 1120112 inodes, 4476416 blocks 223820 blocks (5.00%) reserved for the super user First data block=0 Maximum filesystem blocks=0 137 block groups 32768 blocks per group, 32768 fragments per group 8176 inodes per group Superblock backups stored on blocks: 32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, 2654208, 4096000
10. Create a Filesystem with specific number of Inodes
mkfs provides the facility to create the filesystem with the number of desired inodes. It allows you to override the default number of inodes per filesystem based on the bytes-per-inode ratio as shown below.
# mkfs -t ext3 -v -N 700000 /dev/sda6 mke2fs 1.42 (29-Nov-2011) fs_types for mke2fs.conf resolution: 'ext3' Filesystem label= OS type: Linux Block size=4096 (log=2) Fragment size=4096 (log=2) Stride=0 blocks, Stripe width=0 blocks 701440 inodes, 4476416 blocks 223820 blocks (5.00%) reserved for the super user First data block=0 Maximum filesystem blocks=0 137 block groups 32768 blocks per group, 32768 fragments per group 5120 inodes per group Superblock backups stored on blocks: 32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, 2654208, 4096000 Allocating group tables: done Writing inode tables: done Creating journal (32768 blocks): done Writing superblocks and filesystem accounting information: done
The above can be verified using both tune2fs and df command as shown below.
# tune2fs -l /dev/sda6 | grep -i inode Filesystem features: has_journal ext_attr resize_inode dir_index filetype sparse_super large_file Inode count: 701440 Free inodes: 701429 Inodes per group: 5120 Inode blocks per group: 320 First inode: 11 Inode size: 256 Journal inode: 8 Journal backup: inode blocks # df -i /dev/sda6 Filesystem Inodes IUsed IFree IUse% Mounted on /dev/sda6 701440 11 701429 1% /mnt
My name is Ramesh Natarajan. I will be posting instruction guides, how-to, troubleshooting tips and tricks on Linux, database, hardware, security and web. My focus is to write articles that will either teach you or help you resolve a problem. Read more about
Comments on this entry are closed.
Thanks for the article dude… it is really helpful, thanks for the email from the website that directed me here.
Hi,
Thanks for very usable and helpful article
Hi,
This line is (OK) correct # tune2fs -l /dev//sda6 | grep Inode
but it is better to remove one “/” from /dev//sda6
I am always using:
mkfs.ext4 instead of mkfs -t ext4
and their corresponding versions for different file systems.
What is the difference or what version should be favored?
Thank you for any suggestion.
Best regards,
cee
Many thanks for very informative article.
Please check 8th point entitled “Set the Volume Label for Partition”, seems typo in using appropriate switch (option) for labeling.
Thanks :D.
Thanx for the article, good article, now i know, that i can control my drive with linux with varian of configs. Things that we can’t do explicitly on windows.
Hey Kuldeep,
Thanks for pointing out the typo and corrected the same.
@cee :
Run whereis command for mkfs.ext4,
# whereis mkfs.ext4
mkfs: /sbin/mkfs.ext2 /sbin/mkfs.ntfs /sbin/mkfs.ext4 /sbin/mkfs.vfat /sbin/mkfs.ext3 /sbin/mkfs.ext4dev /sbin/mkfs.minix /sbin/mkfs /sbin/mkfs.msdos /sbin/mkfs.bfs /sbin/mkfs.cramfs /usr/share/man/man8/mkfs.8.gz
mkfs.ext4 is actually a executable file(i.e: filesystem builder) present under /sbin(this path may differ across Linux distrubutions).
when you do “mkfs -t ext4”, the search is on a standard directories list to find out the file filesystem builder(i.e: mkfs.ext4).
Hence both ways are similar and there is no major difference.
Hope I clarified you.
Hi,
I need to know, how to setup the partition inode value as unlimited.
if ex inode limit 100.
+ i created 100 files
+ further i cant create more files. so in this case if set as unlimit it wil be helpful. (or increase dynamically)
can anyone know, pls share it.
Thanks in advance
Thanks Alot
@sathish,
as fas as I know, there is no way of increasing inode value dynamically.
I can give you 2 suggestion on this,
1. When the filesystem is supposed to hold huge number of files, you should actually be thinking of having the partition with large disk size.
2. There is no way of increasing inode count after creating filesystem. You can actually create a softlink for another partition(i.e: partition X) when the inode count is about to go beyond its limit. And the newer files can be continued to store under the softlink. Remember that It is just a logical view(I assume that you are aware of soft/hard link in unix) and the files(under softlink) originally going to reside on partition X.
@Appreciator, Jalal Hajigholamali, Kuldeep, Ghasem Pahlavan, matcet, Azhar Khan :
Thank you so much for your comments.
I am so happy that this article would have helped you in someway 🙂 !!
Thanks to ‘GeekStuff’ and ‘Balakrishnan Mariyappan’ for this artical .
Sorry. This is too complicate for a normal pc-user. Instead tell us where we can find how to make a boot floppy with cfdisk? I have just fixed a floppy with bg-tlb-2.0. But could not find cfdisk on it.
\BgD
After creating the filesystem, I need to press the ENTER key to get the bash again, is there anyway to not have to press the key??
I´m using it in a perl script but it keeps me hanging and I need it to execute the next lines and finish…
🙁
This is old I know, but Im struggling with that right now, thanks in advance