Here's what I did:

OES2 SP1 server (fully patched)
Created an NSS volume on it

I used migfiles command line to copy data from a clustered NetWare NSS volume to the OES NSS volume.

That worked fine.

Here's where it gets odd.

I needed to test some backup stuff. So I ran tape backup against the NSS volume on OES Linux. Backed up okay (but it was a full backup)

Now, since nobody is accessing the data on that volume, my differential backups weren't backing anything up (okay, 3 files that NSS on Linux seems to constantly change on every volume, regardless, but that's it)

So I went into the Migration Utility GUI and setup a project and selected Synchronize

It copied the data over (only overwrite if newer)

Now, it's entirely possibly I missed this or am on crack, BUT:

That was almost 2 weeks ago. However, the backup job for the OES2 NSS volume (and the server has been rebooted several times since then) is always finding a lot of data that "changed". Like 800,000 files or more.

So I'm wondering if the Migration Utility GUI "synchronize" item keeps the file systems synched forever?

Else, how can I explain randomly large amounts of data changing on a volume that nobody uses but the backup software is seeing the data as changed?

Unless of course there's some bug or something with the native Linux Posix stuff on NSS (no TSA registered here).