Is the network share SMB (CIFS) or NFS?  Did you try adding it to Veeam as a repository using one of these methods?  When you do that Veeam imports backups for you if it can read the files.

The illegal character message is usually a corrupted VBM file which is the metadata file for the backup chains.  Here is what you can do to try and fix it -

Why Veeam Shows “Path contains invalid symbols…” When Importing Old Backups

This error typically appears when metadata files (.VBM) inside an old backup chain are corrupted or contain non‑printable characters. Veeam is very strict about XML formatting in VBM files, so even a single bad byte (like 0x02) will cause the import to fail.

From the Veeam Community Hub, this exact issue is described as:

• Veeam fails to import a backup because the VBM file contains an invalid character such as 0x02.

• The error often appears during a repository rescan or when manually importing backups.

🧰 What You Can Do to Fix It

✅ 1. Rename the corrupted VBM file

This is the most common and safe workaround.

Veeam support recommends:

• Rename the .vbm file to something like: backupname.vbm_old

• Then try importing the VBK file directly.

This forces Veeam to rebuild metadata from the VBK file.

This approach is explicitly suggested by Veeam experts in the community thread.

⚠️ Important Notes

• Do not delete the VBM — renaming is safer.

• If the backup is from a newer Veeam version, you can import it, but restores may not be available.

• If the backup is encrypted, you’ll need to specify the password after import.

🧪 Why This Happens

Common causes include:

• Backup files copied from old storage with corruption

• Interrupted transfers

• Bitrot on older disks

• VBM created by a much older Veeam version

• VBM edited or touched by third‑party tools

🛠️ If Renaming Doesn’t Work

If the VBK itself is damaged, Veeam may still fail to import. In that case:

• Try copying the backup chain to a different disk

• Run a repository rescan

• Contact Veeam support (they offer “best effort” support even without a contract)

​@jscooper22 

1. When was the backup you are trying to import made? 
2. And what version of Veeam was used to perform the backup?

Hi all, 

 

1. Rename the corrupted VBM file

This is the most common and safe workaround.

Veeam support recommends:

• Rename the .vbm file to something like: backupname.vbm_old

• Then try importing the VBK file directly.

This forces Veeam to rebuild metadata from the VBK file.

This approach is explicitly suggested by Veeam experts in the community thread.

⚠️ Important Notes

• Do not delete the VBM — renaming is safer.

• If the backup is from a newer Veeam version, you can import it, but restores may not be available.

• If the backup is encrypted, you’ll need to specify the password after import.

Do not do this or suggest it anymore. While this was a valid workaround in the past, it now can create complications that will not be easily resolved.

A support case is the best way to go, even if you’re on community edition. 

Also, the import dialog says “Specify a backup to import. You can import backups from a backup repository or any managed server.” but the dropdown menu only lists the local machine and my two hyper-V hosts, not the repository (which is on a network share).

​@jscooper22 , did you add the SMB share as a repository? Does the VBM file(s) for these backups exist on the share? If so, all you should need to do is rescan the repository to import. 

Mapping network drives won’t work well here due to nuances on how Windows handles this (the share context is only mapped to the user that did the mapping, service accounts won’t have the drive exposed), so adding the SMB share as a repo and importing with rescan is a best bet most likely.

If it won’t go, try a Support case even though it’s community edition.

​@jscooper22  - Just following up on this topic to see if you got an answer and are able to mark the best answer even if it is your answer.  This way the community can see and it will help others.