Welcome to LinuxQuestions.org, a friendly and active Linux Community.
You are currently viewing LQ as a guest. By joining our community you will have the ability to post topics, receive our newsletter, use the advanced search, subscribe to threads and access many other special features. Registration is quick, simple and absolutely free. Join our community today!
Note that registered members see fewer ads, and ContentLink is completely disabled once you log in.
If you have any problems with the registration process or your account login, please contact us. If you need to reset your password, click here.
Having a problem logging in? Please visit this page to clear all LQ-related cookies.
By tylerburns at 2006-03-24 22:59
The /etc/fstab file is used by some programs to determine file system types and mount points. If you are new to linux, configuration files can seem cryptic and intimidating. This tutorial contains an example /etc/fstab file broken down and thoroughly explained. Keep in mind that the sample that is broken down and explained is larger than your typical /etc/fstab so that this tutorial can be applicable to a wider audience. I also included the /etc/fstab from my personal computer.
The numbers at the beginning of each line of the example file are for reference use of this tutorial. If line numbers were added to an actual configuration files the result would be not being able to boot after the next shutdown and the common “mount: can't find /dev/hde in /etc/fstab or /etc/mtab” error message when you try to mount a file system.
The first line of the example file is a comment, which is ignored by the process reading the file. Commenting lines out in linux configuration files is done by using the # character. Comments are very useful for people reading or editing the file. In fact, this whole tutorial could be contained inside a working, fully functional, /etc/fstab file if the lines containing text explaining things were commented out with #. The # character is universally used in linux configuration files to identify comments. Many distributions include a commented line at the beginning of the file to show what each column defines.
As you have more than likely noticed, the /etc/fstab file is divided into columns. The first column defines the device file pointing to the device with the file system to be mounted. The second column is the mount point, or the directory where the file system can be accessed. Column number three defines the file system type used on file system being mounted. An important thing to remember about file system types is that in order for you to be able to mount and use a volume, you must have support for the file system type compiled into the kernel itself or have a kernel module for it. Kernel issues are beyond the scope of this tutorial. The fourth column is used for mount options. Some mount options are file system specific. A chart has been included to explain common mounting options. The fifth column is the dump column. It is used by the dump utility to determine whether or not it should back up the file system being defined. A 0 means it won't be backed up, and a 1 tells dump to back it up. This document will not go into the details of the dump utility. The sixth column is the order fsck checks a file system when the system is booted. A 0 in this column indicates fsck will never check a file system. This value should be used for removable media such as CD and floppy drives. Any value greater than one tells fsck to check the file system at boot. The order they are checked is the numerical order of values assigned in this column.
The second line serves no purpose other than making it easier to read by someone editing the file.
Line number three in this example is technically the only line you absolutely must have, although you should never configure your system this way if you want it to be stable. You should also define a swap partition, which will be explained later. The reason this line is necessary is because it defines the root directory to be mounted when the system boots. Since other file systems are mounted in locations within the root directory, logically you won't be able to mount any file systems without a root directory defined. Mounting another file system before mounting the file system designated to have the root partition is like giving driving directions to use a road that hasn't been built yet.
Line number four defines a separate partition to be used as users home directories. It is often wise to use a separate partition for home directories. It preserves user preferences and files in the case of a new installation or re-installation.
Line number five defines an IDE CDROM drive, and line six defines another optical drive. Typical mount options for optical drives are user, users, noauto, ro, auto, and defaults. Defaults shouldn't be used on workstations or personal computer systems because only root will be able to mount a file system on a CD or DVD.
Line seven is an example of defining a SCSI partition.
Line eight defines a floppy drive. As you probably notice the file system type is auto. This means the system will attempt to automatically detect the file system type the defined device uses.
The ninth of the example file defines a swap partition. While not mandatory, it is highly recommended to define a swap partition in /etc/fstab. You should ALWAYS define a swap partition. Swap partitions are nothing more than dedicated partitions used for virtual memory. It is also possible to use a file dedicated to virtual memory but it is beyond this tutorial. Virtual memory is used when the kernel or applications use up all of your systems RAM or move data there to clear space in the immensely faster system RAM for other data.
The second to last line of the example file shows an NFS mount. As you probably noticed a device file isn't specified. In the column where the device file is normally defined, the NFS server hostname and export directory are specified. The uid=0, and gid=0 tells the computer to make the group and user with the specified ids. In this example both the group and user id are 0, which is the user root and group root. Mounting file systems with specific user and group ids can be a useful security or accessibility tool.
Last but not least, is the proc file system. The proc file system is a virtual file system used by the kernel to store hardware information. Some processes use it to retrieve information about hardware. If left undefined the system will use space on whatever device is mounted as the root partition.
If this tutorial doesn't cover what you are looking for doesn't include what you are looking for, I am going to point you in the right direction on finding help. For file system specific mount options try looking through the mount man and info pages. To access them type man mount or info mount at any shell prompt. Another good source is the Internet. Try searching for a web page using your favorite search engine. Forums such as the one at linuxquestions.org are great places to seek help. Your local bookstore might have some good linux books.
1 device file
2 mount point
3 file system type
4 mount option
5 fsck order
6 dump (1= backup, 0=don't backup)
common mount options:
auto automatically mounts at boot
noauto doesn't automatically mount at boot
owner only root or the owner of the device file can mount this file system
user allows any user to mount a file system, only root or the user who mounted
the file system can unmount it
users same as user except any user can unmount it
ro mounts the file system as read only
defaults mounts the file system with the default options
uid=x mounts the file system as user with id x (ownership)
gid=x mounts the file system as group with id x (ownership)
noexec prevents the execution of files on specified device
Any feedback regarding this document would be appreciated as I plan to write more articles breaking down common configuration files.
The following copyright notice is to keep this document free as in freedom, and price. It is also added to give credit where credit is due.
COPYRIGHT MARCH 2006 by Tyler M. Burns. This document may be distributed electronically. This document may be used as a technical reference for writing other documents. Printing is authorized as long as it isn't published. This documents defines publishing as distribution of this document in exchange for money or goods. This includes but is not limited to books, magazines, and newspapers. Publishing is prohibited unless hard copy, written permission is given by the author to the person or organization wanting to publish this tutorial. Modifications are allowed as long as they are released with this copyright, which must remain unmodified. Modifications require the name and email address of the author of the original document and author(s) of modified version(s) if the modification is made to an already modified version of this document. Modified versions must note the modifications made. Publishing of modified versions require the written permission of both the original author and the author of the modification. Modifications must contain at minimum the email addresses of the author of the original documents and author(s) of modifications in the contact information section below. Failure to provide contact information for modifications will forfeit all rights to the author of the original document. In the event the author and if applicable author(s) of modification(s) do not respond to a request to publish this document, this document may not be published. Appropriate legal actions will be taken for any violation of this copyright.