Testing for missing external files

I use Devonthink to index an archive on a NAS device.
I just wondered if it’s possible to setup a smart group for entries with missing files at the given path?
I had a warning of missing files the other day and it wasn’t clear which entries it was referring to. Although on this occasion I did know about them.

File > Verify & Repair Database… logs all missing items to Windows > Log, afterwards you an use the contextual menu to reveal or trash the items.

This is also covered in the Help > Documentation > Troubleshooting > Repairing a defective database > Missing Files.


I think the issue was indexing my NAS to my Mac and then my iPad via To Go 2. I would move files back and forth to my Mac and offline on my iPad. sometimes same files sometimes different I think this was causing syncing issues. I was getting some of everything duplicates, files I couldn’t sync back to the NAS and Missing files. Perhaps it was just too much of a complex setup?

How would you recommend a NAS <-> MacBook <-> iPad set up. I am wondering if it would be Better to have two databases one for NAS and one for office and iPad and move files between them?

but I will have to keep the links interact …

Is the indexed folder on theNAS being synced by some other application, like a backup utility from the NAS’s manufacturer?

Synology DS716+II
DSM 6.2.3-25426 Update 3

I have a dedicated share/folder and user for Books which is indexed in devonthink the database is on my Mac. The only setting I haven’t tried on the NAS is disabling checksum for advanced data integrity.

I currently don’t run backups or snapshots on this directory

Books Shared folder and file services
I only use SMB file sharing
I do broadcast a dedicated Time Machine folder (with its own user) via Bonjour

SMB advanced settings:
I think locking and durable handles settings are required for Time Machine
I also think Time machine defaults to SMBv3 only and Mac defaults other shares to SMBv3 first theses days?

It’s certainly possible to have separate databases for each purpose.

You could index the same items into two databases.
Also if you duplicate indexed items to another database, no copies will be created in the filesystem.