如何修改文件名合法字符并禁止空格？需日志记录，无需inotify

Hey there, let's break this down properly. Linux doesn't have a single "global switch" to redefine valid filename characters, but we can combine kernel-level mount restrictions, system-wide audit logging, and cross-user environment safeguards to solve this exactly as you asked—prioritizing logging over flaky inotify setups, and handling your mixed non-Linux user base.

1. Kernel-Level Mount Restrictions (Hard Enforcement) #

If your network filesystem (NFS, SMB/CIFS) is mounted on a Linux server, you can enforce valid filename rules directly at the mount layer to block spaces before files even get created:

• For NFS: Use the validname=strict parameter (check your kernel version for compatibility) alongside noacl to avoid conflicts:

  mount -t nfs your-nfs-server:/export/path /mnt/network-share -o validname=strict,noacl
  

  Add this to /etc/fstab to make it persistent across reboots:

  your-nfs-server:/export/path /mnt/network-share nfs defaults,validname=strict,noacl 0 0
  

• For SMB/CIFS: Use validchars to explicitly allow only your approved characters (no spaces) when mounting:

  mount -t cifs //your-smb-server/share /mnt/smb-share -o username=your-user,password=your-pass,validchars=-,_,a-z,A-Z,0-9,.
  

  Persist this in /etc/fstab too:

  //your-smb-server/share /mnt/smb-share cifs defaults,username=your-user,password=your-pass,validchars=-,_,a-z,A-Z,0-9,. 0 0
  

This will immediately reject any file creation/rename with spaces, throwing an error to the user—stopping bad filenames at the source.

2. Audit Logging (Track Violations Without Blocking) #

If you can't enforce hard restrictions right away (e.g., legacy compatibility concerns), use auditd to log every space-containing filename operation for later review:

1. Configure audit rules:
   Create /etc/audit/rules.d/filename-violation.rules with:

   -w /mnt/network-share -p wa -k invalid_filename
   

   Replace /mnt/network-share with your actual filesystem path. This watches for write/attribute change operations and tags them with invalid_filename.

2. Restart auditd to apply rules:

   sudo systemctl restart auditd
   

3. Automate log extraction:
   Make a script /usr/local/bin/audit-filename-spaces.sh to pull out violations:

   #!/bin/bash
   LOG_PATH="/var/log/filename-violations.log"
   # Extract entries where the filename contains a space
   ausearch -k invalid_filename | grep -E 'name=".* .*"' >> "$LOG_PATH"
   # Optional: Clean up old logs to prevent bloat
   find "$LOG_PATH" -mtime +30 -delete
   

   Make it executable:

   sudo chmod +x /usr/local/bin/audit-filename-spaces.sh
   

4. Schedule with cron:
   Add this to your crontab (run sudo crontab -e) to run hourly:

   0 * * * * /usr/local/bin/audit-filename-spaces.sh
   

This gives you a persistent log of every space-containing filename attempt, perfect for identifying repeat non-Linux offenders and building a case for enforcing the hard restrictions later.

3. Cross-User Environment Safeguards #

Since you have non-Linux users (Windows/macOS), don't stop at Linux-side controls:

• Windows/SMB: For Samba shares, add these lines to your smb.conf to block spaces at the share level:

  [your-share]
  vfs objects = fruit streams_xattr
  fruit:invalid_chars = " "
  

  This filters out spaces for Windows users connecting directly to the Samba share.
• User Education: Publish a clear naming standard (e.g., "use camelCase, snake_case, or kebab-case—no spaces") and share anonymized audit logs with your user base to highlight the issue.

Why Skip Inotify? #

You're right to avoid inotify—it's a user-space workaround that's brittle. If the monitoring process crashes, you lose coverage, and it can lag under heavy filesystem load. The mount + auditd approach is system/kernel-level, so it's persistent, efficient, and covers every operation across all users.

内容的提问来源于stack exchange，提问作者puk