9 hours ago, trurl said:

You edited that old post after I had already read it. And since you were still considering using it I assumed you didn't know.

No problem. I really appreciate your help!

BTW (and for anybody that may come across this thread) I found a power supply that has plenty of ports/cables to allow for only (4) per SATA cable: https://www.amazon.com/dp/B0BP8BJ7X4?ref=ppx_yo2ov_dt_b_fed_asin_title&th=1

It has a minimum of (5) SATA ports that can have (4) SATA connections per cable. I believe it comes with three or four of those cables and I will purchase the additional ones I need from Corsair as it is modular. They also have a model that has traditional (not side) connections

Thanks- Bill

As I am rebuilding the drive onto a new one, I am curious if I can reformat or perform another task that will "skip" the bad sectors and allow me to use the drive again by pre-clearing or whatever. Or is this asking for trouble?

It is a 18TB HDD and only two years old. Hard drives are pretty expensive right now too

Sorry, not that well versed in these things.

Edited March 17Mar 17 by wdpaynter

You could try running a preclear cycle, a full disk write may remap the pending sectors, but most likely new ones would appear soon.

41 minutes ago, JorgeB said:

You could try running a preclear cycle, a full disk write may remap the pending sectors, but most likely new ones would appear soon.

Ok, thanks!

So I started rebuilding the drive. It is 18TB and it takes about 2 days to rebuild. Everything was going good until it got to 60-70% completion and the same drives with the power issues began exhibiting read errors (one of the drives had LOTS of errors and growing. The other ones on that same connection had relatively small amount and didn't seem to be increasing. It kept rebuilding and got to about 80% and only 3-4 hours left, and I lost all shares and cannot access the webGUI nor directly switching my monitor to the server (connected via HDMI). Guessing it was the high number of read errors on that one drive? I can here the server making noises similarly to when it was rebuilding

Not sure what to do:

• wait the 3-4 hours and then how do I power down safely (I can;t see anything to initiate it)

• power down now, replace PSU, boot up and start the 2 day rebuild again (if I even can). Again, not sure how to safely power down

• quick press of power button on server case? Makes me nervous

• unplug server power cord and hope my UPS settings work accordingly (doesn't seem likely with everything else not working)

• some other alternative?

If ya'll could provide any guidance, it would be appreciated.

Thanks

Edited March 19Mar 19 by wdpaynter

Post current diags please.

1 hour ago, JorgeB said:

Post current diags please.

I don't know how, as I cannot see the WebGUI nor the command line

Do you have an attached monitor and keyboard?

Just now, trurl said:

Do you have an attached monitor and keyboard?

yes, and a kvm switch from my office desktop. When I click over, I usually see command line, but do everything from the WebGUI on PC. Now there's nothing on either

I've refreshed the webGUI and its back up. Not all my shares. The rebuild of the drive was running, as it shows about 1hr 45mins left.

Are all the read errors with the other drives affecting the rebuild? Or just my she=ares and my ability to run the container/apps?

Hoping I can shutdown after rebuild, install new PSU and boot without further read errors...

26 minutes ago, wdpaynter said:

Are all the read errors with the other drives affecting the rebuild?

It must be able to reliably read all other disks to get the emulated data to write to the rebuild.

2 hours ago, JorgeB said:

Post current diags please.

1 hour ago, trurl said:

  4 hours ago, JorgeB said:

Post current diags please.

Sorry, I was gone for a bit. Diagnostics attached. Rebuild is complete, and disk shows not emulated/disabled, but the notifications point out all the errors.

I will wait for instruction before powering down to install new PSU and cables. Thx!

media-server-diagnostics-20260319-1509.zip

Rebuilt disk12 is mountable and mostly full, but I'm not seeing any shares on that disk so I have to wonder how successfull rebuild was. You filled up your log space with all the errors on other disks so can't see much else about that.

Actually, I'm not sure what filled your log space since syslog in those diagnostics isn't that large but it ran out of space for logging.

What do you get from command line with this?

du -h -d 1 /var/log

and this?

ls -lah /var/log

2 hours ago, trurl said:

Rebuilt disk12 is mountable and mostly full, but I'm not seeing any shares on that disk so I have to wonder how successfull rebuild was. You filled up your log space with all the errors on other disks so can't see much else about that.

My shares disappeared during the rebuild. I thought maybe it was because of all the errors on those other drives.

1 hour ago, trurl said:

Actually, I'm not sure what filled your log space since syslog in those diagnostics isn't that large but it ran out of space for logging.

What do you get from command line with this?

du -h -d 1 /var/log

and this?

ls -lah /var/log

I have never used command line. Can I do it through the WebGUI? b/c I can;t see my unRAID server on my monitor when togging my kvm switch. I was planning to swap PSU after dinner and see how things are. If that solves the read errors, can I rebuild that drive again if it failed?

1 hour ago, wdpaynter said:

I have never used command line. Can I do it through the WebGUI?

5 minutes ago, trurl said:

Well during dinner our neighborhood power went out. It was only out briefly, but server shut down (hoping the UPS did its job).

I guess I'll swap PSU now unless you think I shouldn't

5 minutes ago, wdpaynter said:

power went out

So why log space was full will remain a mystery at least until it happens again.

6 minutes ago, wdpaynter said:

swap PSU now unless you think I shouldn't

I think it would be good to know why the rebuilt disk has no user shares before starting on that.

Start the array in normal (not maintenance) mode and post new diagnostics.

Here is new diagnostics. Oddly, server booted and array started faster than normal. Thx

media-server-diagnostics-20260319-2010.zip

Disk12 has files in 2 user shares, including appdata, which is all over the array. We will worry about that later.

What do you get from command line with this?

ls -lah /mnt/disk12

I will run the command in the morning and report back. Thanks for your help!

9 hours ago, trurl said:

Disk12 has files in 2 user shares, including appdata, which is all over the array. We will worry about that later.

What do you get from command line with this?

ls -lah /mnt/disk12

root@MEDIA-SERVER:~# ls -lah /mnt/disk12

total 0

drwxrwxrwx 4 nobody users 41 Mar 19 20:10 ./

drwxr-xr-x 26 root root 520 Mar 19 20:08 ../

drwxrwxrwx 3 nobody users 19 Jul 15 2024 Mediastorage/

drwxrwxrwx 8 nobody users 146 Nov 7 2024 appdata/

root@MEDIA-SERVER:~#

Regarding doubling up of the shares; Not sure this is relevant (or accurate), but I think sometime early on I read about backing up appdata onto the array and attempted to do so. I beleive I did it incorrectly and someone on this forum (perhaps you or JorgeB) pointed it out. I think we changed the setting(s), but this may be a result. How do I clean it up?

EDIT: when I manually browse the individual disks, I notice that disk12 has (6) container folders in appdata and others have (8)

Edited March 20Mar 20 by wdpaynter

And what do you get from command line with this?

du -h -d 1 /mnt/disk12