So after you reduce size of docker.img, "default shares" will only take about 50G until you add VMs.

I appreciate your help and answering my questions. To be honest, I was a bit sceptical that 2 HDDS could fail - I even thought mobo to blame or cables, but they did. I've ordered a new 4TB drive and it's up and running without any issues. 

Edited May 3, 20251 yr by OlegBudeanu

I still have "Array has 2 utilization warnings" though, one is that one drive is 91% full, but I am not sure how to find the second warning.

Post new diagnostics

Here it is

homeserver-diagnostics-20250504-1536.zip

On 5/3/2025 at 11:58 AM, OlegBudeanu said:

Array has 2 utilization warnings

Is that the exact text? Where do you see that?

7 minutes ago, trurl said:

Is that the exact text? Where do you see that?

here it is, on the dashboard tab

You can set the default warning levels for all disks in Settings - Disk Settings.

 

You can also override those defaults for individual disks in the settings for each disk.

 

Hi again

So after disk replacement and re-build of the array everything was fine for a couple of days. I've downloaded 520gb of data and found out that docker service started to fail to start. I increased the size of the docker slightly, but looks like it's not the case. Can you please help me identify the issue?

Thanks

homeserver-diagnostics-20250508-1707.zip

This is the system log, maybe it will be helpful 

May  8 17:03:13 HomeServer ool www[3766544]: /usr/local/emhttp/plugins/dynamix/scripts/emcmd 'cmdStatus=Apply'
May  8 17:03:13 HomeServer emhttpd: Starting services...
May  8 17:03:13 HomeServer emhttpd: shcmd (211): /etc/rc.d/rc.samba reload
May  8 17:03:13 HomeServer emhttpd: shcmd (214): exportfs -ra
May  8 17:03:13 HomeServer emhttpd: shcmd (216): /etc/rc.d/rc.avahidaemon reload
May  8 17:03:13 HomeServer unassigned.devices: Updating share settings...
May  8 17:03:13 HomeServer unassigned.devices: Share settings updated.
May  8 17:03:14 HomeServer emhttpd: shcmd (226): /usr/local/sbin/mount_image '/mnt/user/system/docker/docker.img' /var/lib/docker 120
May  8 17:03:15 HomeServer root: '/mnt/user/system/docker/docker.img' is in-use, cannot mount
May  8 17:03:15 HomeServer emhttpd: shcmd (226): exit status: 1
May  8 17:03:21 HomeServer webgui: docker stats error: Cannot connect to the Docker daemon at unix:///var/run/docker.sock. Is the docker daemon running?
May  8 17:07:31 HomeServer root: Fix Common Problems Version 2025.03.12
May  8 17:07:33 HomeServer root: Fix Common Problems: Error: Default docker appdata location is not a cache-only share

 

3 minutes ago, OlegBudeanu said:

increased the size of the docker slightly,

Was it filling? Looks like you currently have it at least twice as large as should be needed. Is usage increasing?

5 minutes ago, trurl said:

Was it filling? Looks like you currently have it at least twice as large as should be needed. Is usage increasing?

Not really, I think it was set like that by default and I never edited that. I am not even sure what size it should be or how much the images take

Default 20 is often enough, depending on how many and which dockers you install. I've never needed more but some do.

 

Which disk are you concerned with? They all look fine to me, and no I/O errors that I noticed. I see that was from earlier in thread.

 

Your syslog had a lot of this

May  6 02:47:53 HomeServer shfs: assign_disk: /appdata/Backblaze_Personal_Backup/wine/drive_c/windows/temp No space left on device
May  6 02:47:53 HomeServer shfs: assign_disk: /appdata/Backblaze_Personal_Backup/wine/system.reg No space left on device

many times per seconds a couple of days ago, but it seems to have stopped.

 

Your "default shares" aren't setup the recommended way.

 

https://docs.unraid.net/unraid-os/manual/shares/user-shares/#default-shares

 

Ideally, these shares would have all files on cache or other pool, with none on the array, so Docker/VM performance will not be impacted by slower array reads and especially writes, and so array disks can spin down since these files are always open.

 

 

The Backblaze is a docker which is backing up all my array data into the cloud on Backblaze service. It mounts the array as a drive and backs it up.

 

Using the Fix Common Problems, I set the app data to be cache only, but now it says it is unprotected.

Do I understand correct that I have to set "secondary storage to none" for these shares?

 

 

 

I've moved all the default shares to the cache, as well as the appdata, but still the log says the same

May  8 18:16:07 HomeServer webgui: docker stats error: Cannot connect to the Docker daemon at unix:///var/run/docker.sock. Is the docker daemon running?
May  8 18:16:15 HomeServer ool www[3889414]: /usr/local/emhttp/plugins/dynamix/scripts/emcmd 'cmdStatus=Apply'
May  8 18:16:15 HomeServer emhttpd: Starting services...
May  8 18:16:15 HomeServer emhttpd: shcmd (365): /etc/rc.d/rc.samba reload
May  8 18:16:16 HomeServer emhttpd: shcmd (368): exportfs -ra
May  8 18:16:16 HomeServer emhttpd: shcmd (370): /etc/rc.d/rc.avahidaemon reload
May  8 18:16:16 HomeServer unassigned.devices: Updating share settings...
May  8 18:16:16 HomeServer unassigned.devices: Share settings updated.
May  8 18:16:18 HomeServer monitor_nchan: Stop running nchan processes
May  8 18:16:21 HomeServer ool www[3891131]: /usr/local/emhttp/plugins/dynamix/scripts/emcmd 'cmdStatus=Apply'
May  8 18:16:21 HomeServer emhttpd: Starting services...
May  8 18:16:21 HomeServer emhttpd: shcmd (379): /etc/rc.d/rc.samba reload
May  8 18:16:21 HomeServer emhttpd: shcmd (382): exportfs -ra
May  8 18:16:21 HomeServer emhttpd: shcmd (384): /etc/rc.d/rc.avahidaemon reload
May  8 18:16:21 HomeServer unassigned.devices: Updating share settings...
May  8 18:16:21 HomeServer unassigned.devices: Share settings updated.
May  8 18:16:22 HomeServer emhttpd: shcmd (394): /usr/local/sbin/mount_image '/mnt/user/system/docker/docker.img' /var/lib/docker 50
May  8 18:16:22 HomeServer root: '/mnt/user/system/docker/docker.img' is in-use, cannot mount
May  8 18:16:22 HomeServer emhttpd: shcmd (394): exit status: 1

 

update :

after reboot docker is up and running, but I got a notification from "fix common problems" that

1. these shares are unprotected

2. files exist on the array even though the location is set to "cache only".

Do I understand correctly that if the drive is lost - all the docker information will be lost?

Edited May 8, 20251 yr by OlegBudeanu

Post new diagnostics

Here it is.

Following forums I've installed the CA Appdata backup to backup the dockers on the cache.

And I have the same issue with Docker again - it won't start

 

homeserver-diagnostics-20250509-0809.zip

Edited May 9, 20251 yr by OlegBudeanu

Mover ignores any share that doesn't have a Mover action specified. And nothing can move open files.

 

Go to Settings - Docker and disable.

 

Set appdata, domains, isos, and system shares to:

Primary:cache; Secondary:array; Mover action:array->cache

 

Run Mover, wait for it to complete, post new diagnostics

Hello

So I've done what you proposed - with disabled docker I've set the secondary storage to Array, after that made the mover to move data from array to cache. The Fix Common Problems still says that there is data on the array. But the docker is up and running with reduced image size to 50gb but all the containers are lost.

I've reverted 100gb image back and everything's back. As I understand when I shrink Docker image - it gets deleted ?

Diagnostics attached.

homeserver-diagnostics-20250510-1506.zip

Edited May 10, 20251 yr by OlegBudeanu

Doesn't look like you did anything I asked. Or if you did, you didn't post new diagnostics until after you had already done several things I didn't ask you to do.

On 5/9/2025 at 12:17 PM, trurl said:

Go to Settings - Docker and disable.

 

Set appdata, domains, isos, and system shares to:

Primary:cache; Secondary:array; Mover action:array->cache

 

Run Mover, wait for it to complete, post new diagnostics

Leave those shares exactly as I said.

 

You can leave docker.img alone for now, we can take care of that later.

 

 

Sorry, I did it and reverted and posted the diag.

Sorry, my mistake, here is the new diag after all the actions.

Shares are set to secondary cache and array->cache mover action.

Mover invoked and finished and freed up 500gb cache drive (it was full).

 

homeserver-diagnostics-20250511-1015.zip

appdata and system shares still have files on the array. Mover won't overwrite files, so these are probably duplicates. You will have to clean those up yourself.

 

Probably you can just delete appdata from disk1, and system from disk2, but you should check to make sure the files on cache for these shares are the newest versions.

 

9 minutes ago, trurl said:

appdata and system shares still have files on the array. Mover won't overwrite files, so these are probably duplicates. You will have to clean those up yourself.

 

Probably you can just delete appdata from disk1, and system from disk2, but you should check to make sure the files on cache for these shares are the newest versions.

 

Thanks,

Done

All the info on the array was older than on the cache drive.

Post new diagnostics

here it is

homeserver-diagnostics-20250511-1512.zip