I woke up this morning to an e-mail message from my server: error: error renaming /mnt/user/Downloads/syslog-"Internal IP ADDRESS".log.1 to /mnt/user/Downloads/syslog-192.168.68.57.log.2: Input/output error

Most of my User shares are missing and this is causing issues with several of my docker containers. I think the error above is simply because something couldn't see those shares. No drive/SMART errors are shown.

Posting diagnostics. Any help is appreciated.

media-server-diagnostics-20260716-0739.zip

EDIT: Just stopped array and attempts to restart have failed (array configuration is valid). Going to reboot now that logs were supplied

Edited July 16Jul 16 by wdpaynter

Your cache drive is full and likely caused a host of problems since data could not write to it and also might want to check the filesystem of disk 18.

A reboot appears to have fixed this. Curious as to what may have happened or if my diagnostics/logs point to anything I may have screwed up.

Also, I received a warning that disk18 was low on space and that it is 100% full, but the webGUI shows it has 6+ TB of free space?! I am aware there shows some appdata files on there. This was leftover from a previous settings being incorrect and has since been fixed (can I delete those files?)

Posting new diagnostics after reboot for full drive message troubleshooting

Any help is appreciated.

media-server-diagnostics-20260716-0825.zip

5 minutes ago, MowMdown said:

Your cache drive is full and likely caused a host of problems since data could not write to it and also might want to check the filesystem of disk 18.

I saw those error messages too, but have a minimum set and it never got below 20GB

webGUI of array disks

8 minutes ago, wdpaynter said:

I saw those error messages too, but have a minimum set and it never got below 20GB

You have a minimum cache floor of "0" per your pool settings.

diskFsType="btrfs"
diskUUID="9d37e93b-b771-4d59-aa3b-23aaab01d2ab"
diskAutotrim="on"
diskCompression="off"
diskWarning=""
diskCritical=""
diskExpansion=""
diskShareEnabled="yes"
diskShareFloor="0"   <--- This should be higher than 0 bytes, I personally do 10% minimim pool size
diskBootSize="0"
diskComment=""
diskExport="-"
diskFruit="no"
diskSecurity="public"
diskReadList=""
diskWriteList=""
diskVolsizelimit=""
diskCaseSensitive="auto"
diskExportNFS="-"
diskExportNFSFsid="0"
diskSecurityNFS="public"
diskHostListNFS=""
diskSpindownDelay="-1"
diskSpinupGroup=""
diskFsProfile="raid1"
diskFsWidth="2"
diskFsGroups="1"
diskId="WD_BLACK_SN770_1TB_24036G801555"
diskIdSlot="-"
diskSize="976761560"
diskId.1="WD_BLACK_SN770_1TB_24141V801429"
diskIdSlot.1="-"
diskSize.1="976761560"

Edited July 16Jul 16 by MowMdown

Weird, as I thought I had changed it. Perhaps I only changed it within sabNZBD. Thanks

6 minutes ago, wdpaynter said:

webGUI of array disks

Start the array in maintenance mode, click on disk18 and click on "check filesystem" and see if it reports anything.

Just now, wdpaynter said:

Weird, as I thought I had changed it. Perhaps I only changed it within sabNZBD. Thanks

Note this setting is separate from each share's own minimum free space setting

Edited July 16Jul 16 by MowMdown

got it. thx. Guesing I need to stop the array to change cache minimum? As it is greyed out currently.

Yes

3 hours ago, wdpaynter said:

need to stop the array

which you needed to do anyway so you can restart it in Maintenance mode to check filesystem on disk18

45 minutes ago, trurl said:

which you needed to do anyway so you can restart it in Maintenance mode to check filesystem on disk18

I will do both upon my return. Thank you

Did both this morning. Disk18 check filesystem says: No file system corruption detected.

Thanks

Edited July 17Jul 17 by wdpaynter

1 hour ago, wdpaynter said:

Did both this morning. Disk18 check filesystem says: No file system corruption detected.

Thanks

Can you post new diagnostics, Im curious if the issue resolved itself.

I manually removed the cache files that were on disk18 prior to running check filesystem. I feel like something I run keeps putting some on there despite settings being corrected and all other appdata/cache files being handled correctly. Originally, the intent was to backup the appdata/cache to the array as I followed instructions somewhere to do so. I was later told that this was wrong. Not moving the originals from cache to array, but copy/backup. I guess having a NVME pool likely negates the need anyway

media-server-diagnostics-20260717-0938.zip

Edited July 17Jul 17 by wdpaynter

 Everything looks good from what I can see, you only have 1 share that mover moves things from your cache pool to the array now, whatever your "M--------e" share is

appdata                           shareUseCache="prefer"  # Share exists on cache
domains                           shareUseCache="prefer"  # Share exists on cache
D-------s                         shareUseCache="only"    # Share exists on cache
isos                              shareUseCache="only"    # Share exists on cache
M----------e                      shareUseCache="yes"     # Share exists on cache, disk1, disk2, disk3, disk4, disk5, disk6, disk7, disk8, disk9, disk10, disk11, disk12, disk13, disk14, disk15, disk16, disk17, disk18
system                            shareUseCache="prefer"  # Share exists on cache

The only thing that looks sus is the split level on the M share, make sure its also set to something, the config shows it as blank

2 hours ago, MowMdown said:

The only thing that looks sus is the split level on the M share, make sure its also set to something, the config shows it as blank

I am not familiar with this setting and am not very good with unRAID in general. Any recommendations? I don't have deep folder hierarchy but understand that I should likely have it set. It looks to be set to "Automatic"

Thanks for your help!

Edited July 17Jul 17 by wdpaynter

That should be fine. Split level just tries to keep files together on the same disk if they belong together (are in the same folder). Highwater allocation often has the same result, where files that are written at the same time likely will end up on the same disk. The only penalty for files that belong together ending up on different disks is a delay when spinning up another disk. And since split level takes precedence over allocatiion and Minimum free, there can be some downside to making it keep files on the same disk, where it might try to put files on the same disk even if they won't fit.

One last question. At some point disk18 was incorrectly reporting its free space and it looks to be corrected. It only has 6GB of free space, which seems really low. Should I move files to another disk manually (Krusader)? Or leave it be? Pic of drives attached. Thx

It's a good idea to keep some space free on each drive in case filesystem repair needs some space to work in. Disk5 also doesn't have much free space.

18 minutes ago, trurl said:

It's a good idea to keep some space free on each drive in case filesystem repair needs some space to work in. Disk5 also doesn't have much free space.

Should I move files to other drives manually?

You don't need krusader, and if krusader is configured to allow you to access individual disks, you must be sure you don't mix disks and user shares when moving/copying.

Built-in File Manager won't let you make that mistake.

I was just going to move them and then let unRAID do its thing moving forward (not manually move). Not sure excatly how this happened in the first place, as I have it set to a 97.7GB minimum

I did first one using Krusader and second using File Manager. Will use File Manager moving forward. Thx! Maybe i moved a share to a disk previously or vice versa that screwed things up to begin with?