Basics of Linux File System

Directory

• It is a special type of file in a Linux file-system.
• It acts as a container for other files and directories.
• The content stored in a directory is essentially a list of Directory Entry.
• These are organized in hierarchical structure.
• In Linux everything is a file.

Directory Entry

• It is a record in a directory.
• It associates or maps a file name(human readable) with its inode number in a file-system.
• A directory entry contains the file name and pointer to the file's inode.
• It is stored in Directory.
• It doesn't store the file metadata that is done by Inode(Index Node)

Inode(Index Node)

• Inode is a date structure.
• It stores the metadata and other information of files except the file name.
• Each file and Directory has its own unique inode.

Type of Info stored in Inode

1. Metadata

• File type
• File permissions
• Owner and Group ID(UID & GID)
• File size
• Timestamps (birth, access, modification, change times)
• Number of Hard link & Soft Link (Symbolic Link).

2. Location of Data blocks

Pointers of the disk blocks where data is stored.

the file name and the directory path is not stored in inode these info are managed separately by Directory Entry that map file names to inode numbers.

Working of Inode

• Each file and directory has its inode.
• When new file or directory is created the file-system assigns a unused inode to it.
• It serves as identifier for file and directory.

Viewing Inode

ls -i

output ->
4073139 bubble  4073136 bubble.c  4073141 bubble.png  4073138 insertion  4073137 insertion.c

This will return the file names with it's inode numbers.

Creation of inodes

• When a file system is created a fixed number of inodes are pre-allocated based on the size of file-system.
• The number of inodes are in the ratio with the file system size. (like 1 inode per 16KB).

Dynamic assignment of inodes

• When new file or directory is created the file-system looks for first available unallocated inode from the inode table and assign it.
• inode number can be reused, when file or directories are deleted the inode of those are marked unallocated by the file-system and then can be reused.

Hard link

• Hard link is a directory entry that maps the file name with an inode in a file-system.
• A hard link creates a additional name for the same file, pointing at the same inode thus referring the same data.
• Hard links are indistinguishable from the original file even the timestamps will be the same(access time, modify time etc).
• Each file must have atleast one hard link because the hard link maps the file name with its inode which contains the metadata and pointers to the file content.
• Hard links can't be created between different file systems or partitions.

Creating a hard link

touch hello.txt
ln hello.txt world.txt

• Here the world.txt is the hard link to hello.txt, they both are pointing to the same inode number thus the same data.
• Both files now contains the same data changes done to one will reflect in other file.
• Deleting one doesn't affect the other.