How to Delete Files with Problematic Names in Linux: 6 Proven Methods

Linux file naming rules

In Linux a file or directory name can contain any ASCII characters except '/' and the null byte. Although many characters are allowed, using special symbols (e.g., '-', '<', '>', '*', '!') is discouraged because they can interfere with command parsing.

Delete by specifying the path

If a file name starts with a dash, rm -static is interpreted as an option and fails. Prefix the file with ./ to treat it as a pathname:

$ rm ./-static

Delete using the "--" separator

The double‑dash tells rm that everything following it is a file name, not an option:

$ rm -- -static

Delete by quoting the name

Wrap the problematic name in quotes to prevent the shell from interpreting special characters: $ rm "<>\!*" Note that this does not work for names that are themselves shell metacharacters, such as !*.

Delete by escaping special characters

Prepend a backslash to each special character to neutralize its meaning: $ rm \!* Escaping also works for leading spaces, e.g., $ rm \ abc removes a file whose name begins with a space.

Delete by inode number

When a filename is unreadable or contains garbled bytes, locate its inode with ls -i and delete via find:

$ ls -i
1703907 corrupted_name.pdf
$ find . -inum 1703907 -exec rm {} \;

The inode uniquely identifies the file, independent of its name.

Delete using wildcards

For groups of files sharing a pattern, a wildcard can be used, but caution is required to avoid accidental mass deletion:

$ rm *.pdf

Summary

Delete by specifying the path (./filename)

Delete with the "--" separator

Delete by quoting the filename

Delete by escaping special characters

Delete by inode number

Delete using wildcards (with care)

Choose the method that best matches the type of special filename you need to remove.