Disk Churning

Your original thought about spotlight indexing is the issue. The index probably got damaged and it is continually reindexing. Disable Spotlight (exempt) on this volume, restart. the churning should be stopped. Then you can remove the exempt spotlight and after a while, the indexing should stop and you should be back to normal.

Thanks.

I've done this now and am waiting for the new indexing to stop...

So far, today anyway. MUCH less kerchunking from the RAID drive system.

Thank you!

@bkdad 

Just remember, SoftRAID's driver does not "generate" any IO. Only the app can do that (verify disk for example). So anything going on like you describe is MacOS.

Well, a funny thing happened.

The Thunderbay kept chugging away, although at a reduced rate than before.  (Thanks for the tip!)

Then, I updated Sequoia yesterday to the latest drop from, umm, yesterday.

Guess what!?  Much reduced disk access now.  Yes, I know that it's only been a day or so, but it's definitely noticeable. 

Go figure.

@bkdad 
I just had a volume churning, it was of course Spotlight. Disable it, goes away. Remove the Exemption, it slowly builds a new index, then finally stops.
Spotlight seems to have indexing issues on volumes with lots of files.

This problem keeps emerging.  Over and over.  The fix works, but only for a week or so.  It's a nuisance.  Clearly a long standing macOS problem.  My guess is that if it hasn't been resolved by now, it will never be resolved.  Apple's engineers are too busy being directed to make new emojis and difficult to look at user interfaces to fix bugs.  No glory in that.

Yesterday, I got fed up and excluded all the volumes in the external RAID drive from Spotlight.  Silence!  Now, I'm learning the virtues and subtleties of EasyFind.

This morning I got thinking about it and wondered if anybody has written a script that is launched as a cron or launchd job that indexes the volumes of choice at a time of choice and once the indexing is complete then turns off indexing for those volumes until the next time.  That would allow for having the external volumes be indexed at 3AM, for example, every morning and left quiet for the rest of the time.  

I know the functionality is available with mdutil and the rest, and I could probably learn to do it all myself, but if somebody has already done all the work for this and debugged it, why reinvent the wheel?

@bkdad 

Apparently this kind of script is very common in post production environments.
ChatGPT can help you write a custom script in a few minutes. 
I don't know if one exists that is "open source" or not.}
You can try Github. But this looks trivial for ChatGPT, especially since you haev some terminal experience.
Just ask it using a clear description like you did in your post and it comes up with concepts, step by step and actual script code.
then you just have to put your variables in and debug it. Probably a 10 minute job for you.

@bkdad 
BTW: One reason this happens is unclean shutdowns, and when disks are "ejected" without being unmounted.

Does your system suffer from your drives being ejected randomly? That used to be a common Thunderbolt problem, which has mostly resolved with newer Thunderbolt versions, cables/connections, etc.

Sorry for the late response.  Holidays, etc.  (No excuse, though.)

I haven't had any problems at all with random drive ejections.  Perhaps once a year I'll pull out an SD containing photos prematurely and get that warning.  Maybe not even that often.  The hardware seems to work exactly as advertised and expected.  I wish it was that simple.

Thanks for the suggestion of investigating ChatGPT.  I will do just that.  Might be fun.

I find that I do not miss Spotlight...