The only thing that shows up is that the data is apparently still there since the space on the drives is the same.
There were no warnings or errors prior to this, so I am really confused and in more than a small bind as I don't even remember all the share names or all the tools and settings for things like cloudflared and dockersoctet and ... too many other things.

The hard reboot wiped the logs I suppose. I will see if I can get it to go online long enough to pull them so I can share.

Apologies if this sounds dramatic, but to me this is a fairly massive failure and I am unsure how to proceed so I don't lose years of data.

tower-diagnostics-20260410-1835.zip

Edited April 11Apr 11 by LordShad0w
added Diags

Boot in SAFE mode and post new diagnostics

It's currently running the parity check. Also, how can I force it to boot into safemode if I can not get it to ever stop the containers and reboot?
I appreciate your help.
This whole issue started because I asked it to stop before I was going to issue the shutdown command and it just never... stopped.

It just said it was retyrying stopping everything for over 45 minutes before I did a hard shut down.

ETA I did just recently update from 7.2.3 and had no reason to reboot or anything since. Not sure if that matters but thought I would share.

Edited April 11Apr 11 by LordShad0w

Disks and pools are mounted, user shares are not. But something is trying to access user shares, perhaps too early. Perhaps some rogue plugin, you have a large number of plugins.

So should I start nuking plugins? I have no problem doing that

Reboot in SAFE mode so no plugins are loaded and see if that helps

Just now, trurl said:

Reboot in SAFE mode so no plugins are loaded and see if that helps

I was asking you how I can do that, since the machine refuses to ever stop so that I can proceed to the reboot in safemode step?

Power switch

Bro...

I am neither trying to be rude nor obstinate in any way and am genuinely asking for help.

Anyhow, thank you for at least responding I guess.

I had to pull out a monitor and kb to plug in directly to the server, but I was able to get into safe mode. Here are the files.

Thank you.

tower-safemode-diagnostics-20260410-2016.zip

Diags after array start please.

The safemode ones were the after start. If I normal start it again it won't let me shut it down again. Is that ok?

I am starting it now. Should have diags in about 10mins

Edited April 11Apr 11 by LordShad0w

The array is not started on the safe mode diags, diags in safe mode with array started and in normal mode, not maintenance mode.

oh ok, i didnt understand that. I will do it now. Thank you Jorge

@JorgeB Should i click maintenance mode so it doesn't spin up disks or just hit start? (It is in safe mode presently) I will wait for your answer before proceeding so I do it correctly.

tower-safemode-diagnostics-20260410-2349.zipHere you are.

Users shares appear to be OK. Confirm if all is working correctly, and if it works in safe mode and not normal mode, it suggests a plugin is the issue.

In safe mode, it shows my shares, but now only as disk shares but not user shares as they were, somehow?
all my dockers are gone still.
I removed any plugins added in the last week.

Not sure how to proceed. ☹️

I removed the plugins added in the last few days and now things seem ok. I am having to rebuild databases and rescan everything and recreate my plex, but at least it works again. Thank you!

I removed the plugins and now everything is rebuilding. Gonna take a few days but at least I am up and running again. Thank you so much.

So, while things are working again, I have a major problem. Aside from about 20 shows, all the others in my TV share are gone. When I calculate the size of the share it comes in at just over 4TB. I have tried to see if they got moved to another disk somehow, but no joy.
It is odd since it shows the 4TB but when I look at the array numbers, the 16TB drive that had the bulk of my tv shows went from 65% filled down to 49%.

I don't see how that much data just vanished into thin air.
Did you guys see anything that would indicate an operation that might account for this happening when you looked over the logs?

Thanks guys.

Edited April 12Apr 12 by LordShad0w

Post new diagnostics

Sure, give me a sec

tower-diagnostics-20260412-1833.zip

Edited April 13Apr 13 by LordShad0w

Your syslog has btrfs csum errors on cache. This is almost certainly bad RAM.

Don't use your server any further until you have tested and fixed RAM.

Not necessarily related, but your system share is all over the array. It is the most important share to have on a pool outside the array, like your appdata is.

Why do you have 100G docker.img?

We can work on those things after you have good RAM.