Mastering mount --bind: Seamless Directory Redirection on Linux

During Linux system administration, storage directories can fill up unexpectedly, and expanding the original disk may be impossible. A common quick fix is to create symbolic links, but multi‑level symlinks become confusing and error‑prone. This guide introduces mount --bind as a cleaner alternative.

What is mount --bind ?

The mount command is familiar to sysadmins. Using the --bind option, you can remount an existing directory onto another location, making the target directory a mirror of the source. All accesses to the target are transparently redirected to the source.

Basic Example

First, create two directories and place different files in each:

mkdir /tmp/test1
mkdir /tmp/test2
echo "data in test1" > /tmp/test1/file1.txt
echo "data in test2" > /tmp/test2/file2.txt

Check their inodes to see they are distinct:

ls -i /tmp/test1
ls -i /tmp/test2

Now bind /tmp/test1 onto /tmp/test2: mount --bind /tmp/test1 /tmp/test2 After the bind, the inode numbers and file contents under /tmp/test2 match those of /tmp/test1. Any modification made through /tmp/test2 actually changes the files in /tmp/test1.

Creating a new file in /tmp/test2 also creates it in /tmp/test1: echo "new file" > /tmp/test2/new.txt Listing both directories shows the same files, confirming the bind relationship.

How It Works

When mount --bind /tmp/test1 /tmp/test2 runs, the kernel hides the original entry of /tmp/test2 and records a mapping in a VFS vfsmount object. This object stores a hash‑based relationship between the source and target paths. Subsequent accesses to /tmp/test2 are redirected to the inode of /tmp/test1, while the underlying data remains unchanged.

Persistence

The bind relationship exists only in memory; a reboot clears it. To make it permanent, add an entry to /etc/fstab:

/tmp/test1  /tmp/test2  none  bind  0 0

Advanced Uses

Temporary configuration testing : Create a test config file in /tmp and bind‑mount it over the real config path. Run the application to verify behavior, then unmount to revert without touching the original file.

Read‑only bind mounts : Provide read‑only access to a directory while keeping the source writable. Use the ro option: mount --bind,ro /path/source /path/readonly This is useful for giving developers or auditors view‑only access to configuration files.

These examples illustrate the flexibility of mount --bind for managing storage constraints, testing scenarios, and access control without the pitfalls of deep symbolic‑link chains.

Conclusion

mount --bind

offers a reliable way to redirect directory accesses, preserve inode consistency, and maintain clear, reversible mappings. Remember to persist bindings in /etc/fstab if you need them across reboots, and explore the advanced options for temporary or read‑only mounts.