inodes

Inodes are data structures in Unix-like file systems that store metadata about files and directories, excluding their names and actual data blocks. Each inode has a unique inode number within the filesystem.

Contents typically include the file type and mode, owner and group IDs, file size, access time, modification time, change time, and a link count. They also contain pointers to the disk blocks that hold the file’s data. Block pointers usually include direct pointers and indirect pointers (single, double, and triple) to access additional blocks. The exact layout depends on the filesystem (for example, ext2/3/4 use a combination of direct and indirect pointers).

Directory entries map filenames to inode numbers. An inode can have multiple hard links; the link count tracks how many directory entries reference it. When a link is removed, the count decreases; when it reaches zero, the filesystem may reclaim the inode and its associated data blocks.

Allocation and management: Inodes are allocated when a filesystem is created, and there is a fixed inode table that determines the maximum number of files the filesystem can hold. If a filesystem runs out of inodes, new files cannot be created even if there is free data space. Inodes are part of the filesystem metadata and are checked and repaired by filesystem tools such as fsck in case of corruption.

Tools and usage: Inodes can be inspected with commands such as ls -i to display inode numbers, stat to view detailed inode fields, and df -i to show inode usage across mounted filesystems. The inode concept is central to how Unix-like systems locate and manage file data, separating metadata from the directory-level naming information.