Last minute addition: The 17.00GB backup on Alice (2012 Intel MacBook Retina) just completed, and appears to be OK, although the CIFS connection was corrupt to the point I had to clobber it and reconnect to see anything on the Synology box. Also, I got the history backwards: the oldest Mac, a 2009 table top machine, spontaneously and seamlessly converted its backup to a .sparsebundle following its last update, after which it no longer participated in the latest new code game; it's the 2012 MacBook that still persists in calling its file a .backupbundle, despite the need to twice be re-created (after hiding the original).
Now, that's strange behavior for a world that is, despite all appearances, still decidedly deterministic at its heart.
Hoorah, Semper Fi, Next Bug, please...
---------------
OK, on my 13" MacBook Pro M1 2020 (MacBookPro17,1) Monterey 12.0.1 backing up to Synology DiskStation DS-215J the problem exists, but the result is a bit different - it completes the backups, then displays the dates from the last successful backup, about there weeks ago (before the upgrade to 12.0.1, Duh!) along with that tiny information "i" symbol, which when clicked proclaims the backup couldn't finish, but will complete when my Mac is unlocked.
My Mac? Unlocked? What country do you think this is, sir?
Turns out the 'Mac' the message refers to is the AFP connection on the Synology, which isn't locked in any way trivially obvious: connections look good, latest system update Just In Case, nothing changed, lots of space available.
I haven't bothered looking for hidden files just yet, I'll check that in a moment, but nothing about the way that the MacOS stages it's backup data is very different from what has been around for years, so no surprise if you've been out in the industry watching (and doing) this stuff for a lot of years.
I can verify that there seems to be something a bit off all the way around, as sitting alongside I have my (decidedly Intel based) MacBook Pro Retina Mid 2012 running the latest update of Last Gasp Catalina (10.15.7, loaded today at the same time as the Monterey 12.0.1 on the M1) and the backup behavior is definitely affected, although not in the same way.
The complete informational message on the M1 machine is:
Time Machine couldn't complete
the backup to "Darwin.local"
Time Machine did not finish backing up
because some files were unavailable.
Backups will resume when your Mac is
unlocked.
I don't currently know what the Intel machine will say, as it spontaneously shed its CIFS connection on Darwin (the Synology box, if that wasn't obvious) about the same time it decided to start a more-or-less complete backup of exactly 17.00GB, which is halfway complete byte count wise; this is a significant change in behavior from the previous several hours of fooling around with these two machines, what a way to spend a day off, eh?
Other stuff of interest: similar bugs have appeared previously, extinguished via various means; the backup format changed from .backupbundle to .sparsebundle a ways back, except for Kay's aging 2009 table Mac, still useful with all the extra memory we crammed into it, haven't even bothered to put the 1/2TB SSD in it yet; twice since that change, which trashed the easy utility of a lot of backup history for the Retina MacBook, I've had to 'hide' the previous backup file in a stash directory and let TM begin again, which is what it looks like it's decided to do a third time for the Intel machine; we have occasionally had weird behavior that disappears when we disconnect our VPN service, not that surprising (but Hey, who's looking at what we're transferring on an internal network connection? Good question...) and a few other glitches I'm sure everybody has encountered.
More as it happens - there's bound to be a bit of action on this over the next small while. --TR