A mount point is a directory or file that provides access to a new file system, directory, or file.
The mount point is used to implement the ability to dynamically attach / detach disk partitions to the file system while the operating system is running. Support for this feature in UNIX-like operating systems based on monolithic kernels is provided by the kernel of the system and the system program, which is invoked using the command line utility mount . Mounting storage devices is used in any Unix-like operating system , in particular, the mount utility is present in Version 5 AT&T UNIX.
To attach the second partition of the first hard disk to the directory tree on a UNIX-like operating system (for example, GNU / Linux ), you must run the following command at the command prompt:
mount -t ext3 -o iocharset = koi8-r, codepage = 866 / dev / sda2 / mnt / disk2
mount / dev / sda2 / mnt / disk2
This command, turning to the kernel , will attach ( mount ) the second partition of the first hard drive to a directory with the name / mnt / disk2, and the mount point can have any name. To do this, a device file named sda2, which is a file for inputting and outputting data to a hard disk driver, after mounting in the specified directory, will output data to the ext3 file system driver, and then, through the virtual file system implemented in the kernel, the mounted file system will be presented in the mount directory. For user programs, it will look like the attached file system is part of the root file system. In this example, the mounted device represents the second partition ( 2 ) of the hard disk ( d , for optical drives - c ) of the first channel of the controller ( a ) and uses the SCSI command set (this is indicated by the initial letter s ).
Also, the mount command when mounting will proceed from the value received from the user (usually the superuser has the privilege of mounting) the parameters with the -t switches: in this example, the ext3 partitioning is used to indicate the partition type of the disk partition, and the -o switch is used to indicate the options passed to the command mount . In the first example, the code page number 866 and the encoding type KOI8-R were passed to the mount command as parameters. If the parameters do not match the actual properties of the mounted disk partition, the user will receive an error message. The second example is a simplified use case for this command, in which all actions will be performed automatically based on the default settings. You can unmount a disk partition with the command:
umount / dev / sda2
umount / mnt / disk2
Moreover, both options are equivalent in their action and result. You should also pay attention to the spelling of the command: it is spelled correctly umount , not unmount.
In next-generation Unix-like OSs, such as Hurd (GNU / Hurd), which runs on top of the microkernel , mounting file systems, as presented in regular UNIX-like OSs, has been replaced with technology that uses system servers. In Hurd OS terminology, they are called translators . Translators in the Hurd OS provide much more functionality than is possible with conventional UNIX-like OSs.
Windows Mount Points
On Windows, mount points are available using the NTFS file system. Mount points are implemented through a more general reprocessing point technology  . There are two types of mount points: the mount point of the directory ( English junction point ) and the mount point of the volume ( English volume mount point ). Mount points of the first type are created through the console command mklink / J , and mount points of the second type are created through the mountvol command.