I decided to add an additional SSD to my server. I watched Spaceinvader's guide to make sure I didn't break anything. ( I am a server noob) I got the new SSD installed. My original SSD was a cache and contained my Appdata, Domains, Iso, System, and Nextcloud folders. I decided I wanted my second SSD to be a second pool so they looked like this...

 

Original SSD: Domains

New SSD: Appdata, Iso, System and Nextcloud.

 

I rename the SSDs and set each folder to its preferred pool. I knew a lot of data would need to transfer from Orginal SSD to new SSD. So I went ahead and ran Mover.....while my dockers were running...

 

My data didn't move and I was confused why, found a reddit post that said I needed to turn off my dockers first. I went to the docker tab to stop them and they are all gone. I freaked out and went to settings and turned of docker all together and then ran mover again. My data is currently moving and I am just sitting here knowing that my dockers aren't going to magically come back after this move.

 

 How screwed am I? Is there a way to reverse this damage or I am i going to have to redo every one of my dockers? Thanks for any help in advance. I am sad but hopeful.

 

 

server-diagnostics-20220420-2303.zip

Edited April 21, 20224 yr by Anthadeas
Added Diagnostics

Mover won't move files between pools. It will move cache-yes shares from the pool designated for the share to the array, or move cache-prefer shares from the array to the pool designated for the share.

 

Nothing can move or delete open files. Stopping each docker would allow appdata to be moved, but system can't be moved unless you go to Settings and disable Docker (which would stop all dockers) and VM Manager (which would stop all VMs), since the system share contains docker.img which is open by the Docker service, and libvirt.img, which is open by VM Manager service.

 

According to your Diagnostics, you now have Docker disabled, but VM Manager is still enabled.

 

Your appdata is all on cache_vm. You have it set to prefer cache_docker_system, but as mentioned, mover won't move between pools. Mover only moves cache-yes shares from the designated pool to the array, and moves cache-prefer shares from the array to the designated pool. Mover ignores cache-only and cache-no shares, and any pool except the pool designated for the share.

 

Your system share is set to prefer cache_docker_system, but it also has files on cache_vm and disk1.

 

Not sure why your dockers would be gone, except of course, docker is currently disabled. Maybe you left out something in your description. In any case, your containers can be reinstalled exactly as they were using the Previous Apps feature on the Apps page should that be necessary.

 

Simplest (and what I would do) would be to do these moves yourself using Midnight Commander (google it, mc from the command line). Just be sure you only work with the disks / pools when doing these moves, and not the user shares since the user shares are just another view of the disks / pools, and you could lose data by doing a move where the source and destination are the same.

 

Or safer, you could use Dynamix File Manager plugin if you were running Unraid v6.10rc3+.

 

You can get mover to do them for you but it would require several moves with changing settings for the shares so they would get moved to the array and then moved to the designated pool.

 

Also, mover won't move duplicates, so even if you have everything set correctly to get one of these mover steps done, some files might not move if they already exist on the destination.

 

You can see how much each user share uses on disks / pools by clicking Compute... for the share on the User Shares page, or the Compute All button.

 

Probably we will have to work this out in several steps with mover, but I am going to sleep now. See if you can make any sense of what I said above and we can work on it tomorrow.

 

 

trurl you are extremely helpful.

 

First thing I did was just trying to reinstall the dockers from the Previous Apps page. They all reinstalled and started but the Webui button was gone (Nextcloud, Photoprism) and I got error 404'd when trying the IP address. I just wanted to see if it worked out of the gate but I am assuming since i renamed the pools they will probably stay broken until I get the data sorted out correctly.

 

I went ahead and updated Unraid to v6.10rc3+ so I could install Dynamix File Manager. I have to head to work soon so I went ahead and tried your last option of moving everything to the array and then back to the correct pools using mover. (There isn't too much data on my pools yet)  I don't think anything really moved. So I will try Dynamix or Midnight Commander as soon as possible (Busy schedule over the next couple of days)

 

Question for you in the mean time. If I finally get all of my files sorted out correctly between the two pools will my dockers 'just work' again? Or will have some real work ahead of me? My Diagnostics are downloading now. I'll edit as soon as it is finished

1 hour ago, Anthadeas said:

moving everything to the array and then back to the correct pools using mover.

To get a share moved to the array from whatever pool it is on, the share would have to designate that pool and be set to cache-yes. Then to get it moved to another pool the share would have to designate that other pool and be set to cache-prefer. And of course, nothing can move open files and mover won't move duplicates.

12 hours ago, trurl said:

You can see how much each user share uses on disks / pools by clicking Compute... for the share on the User Shares page, or the Compute All button.

 

1 hour ago, Anthadeas said:

If I finally get all of my files sorted out correctly between the two pools will my dockers 'just work' again?

Your pools are part of user shares, so if you have only specified user share paths in any of the configurations it should be able to find the files whichever pool they are on. If you have specified a path to a pool then you would have to update that.

 

1 hour ago, Anthadeas said:

i renamed the pools

Probably the shares are designating pools that don't exist since they have been renamed.

 

Diagnostics might clarify what is happening. Also since you have upgraded the newer diagnostics make it easier to see how each share is configured and where they have files.

Quote

To get a share moved to the array from whatever pool it is on, the share would have to designate that pool and be set to cache-yes. Then to get it moved to another pool the share would have to designate that other pool and be set to cache-prefer. And of course, nothing can move open files and mover won't move duplicates.

Yep that is what I tried to do. I made sure that I disabled VMs and Docker. But It appears nothing moved.

 

Quote

You can see how much each user share uses on disks / pools by clicking Compute... for the share on the User Shares page, or the Compute All button.

Oh that is awesome. That helped see what was going on much easier. I also got Dynamix File Manager figured out a little bit so I think I will try and moved some stuff around and see what happens.

 

Quote

Your pools are part of user shares, so if you have only specified user share paths in any of the configurations it should be able to find the files whichever pool they are on. If you have specified a path to a pool then you would have to update that.

Right, I  checked and mine are setup as user shares. I think what is happening is that there is now a copy of some of the appdata folder in both cache drives now. the copied one has a nextcloud folder but nothing in it. I bet it is directing to that one. I think I will try and delete that one and see what happens.

 

I am trying to link you new logs I started downloading them this morning when I went to work. I OpenVPNed in at lunch and they were finished. But something messed up and I couldn't save them to a folder. I started downloading them again a few hours ago. They are still downloading. (Not sure why it is taking so long, but it appears to be going through every file and photo on my Nextcloud and Photoprism now that my Docker is turned on) I will upload them as soon as they finish. By the way I tried looking at logs since they help you so much but I feel like I am just looking at 1000 lines of gibberish lol. Thanks again for all of your help so far.

I seldom look at logs until I have looked at other things in diagnostics. Those other things help suggest what might be useful in logs, a lot of it doesn't make sense to me either but only some of it is typically useful.

 

Diagnostics should only take a few minutes or less to create and download, and they are zipped so are typically 100 KB or so in size. As you can see your earlier diagnostics were less than that. Don't know what might be going on with your remote access.

12 minutes ago, Anthadeas said:

it appears to be going through every file and photo

Diagnostics doesn't go through any of your files except config files on flash.

Just now, trurl said:

Diagnostics doesn't go through any of your files except config files on flash.

It does get the results of some command lines such as df, these can be seen in the system folder in the diagnostic zip. And some other things like the Compute... for each share. Still shouldn't take long or be large.

10 minutes ago, trurl said:

It does get the results of some command lines such as df, these can be seen in the system folder in the diagnostic zip. And some other things like the Compute... for each share. Still shouldn't take long or be large.

Right, and I remember the first one being super fast too. but I have been downloading this one since I have been home. and it has been cache_vm\/Nextcloud/Updater etc etc etc for hours. I really think it is going through every single thing. If i cancel it and do it with docker disabled it will be done in seconds. Edit: Diagnostics just finished! Good timing I guess.

 

server-diagnostics-20220421-1906.zip

Edited April 22, 20224 yr by Anthadeas
Added Diagnostics

If you look in those diagnostics under the shares folder, there is a file called shareDisks.txt. That is a new diagnostics file in 6.10 that summarizes the settings and disk/pool usage for all shares. Earlier versions required us to look at each share .cfg to get that.

 

And you can see that you have no shares set to cache-yes.

20 hours ago, trurl said:

Mover only moves cache-yes shares from the designated pool to the array, and moves cache-prefer shares from the array to the designated pool.

So nothing is going to be moved to the array since nothing is cache-yes.

 

Typically appdata, domains, system shares are cache-prefer (or only) but it is OK to have others prefer to stay on cache if that is really what you want for some specific uses. But this doesn't seem to be what you want since it is what you are trying to fix now.

And your appdata and system shares are on both pools.

1 minute ago, trurl said:

 

 

And you can see that you have no shares set to cache-yes.

So nothing is going to be moved to the array since nothing is cache-yes.

 

 Right, I see what you saying. I did set them all to cache-yes and ran mover. I didn't see anything move but I assumed i was just dumb and maybe it did. So then I set them all to the Cache_etc I wanted to see if something did move from the array would move back. But nothing moved. I just haven't changed them since.

 

Quote

Typically appdata, domains, system shares are cache-prefer (or only) but it is OK to have others prefer to stay on cache if that is really what you want for some specific uses. But this doesn't seem to be what you want since it is what you are trying to fix now.

Yeah I see what your saying. I do want those all on my cache. I just wanted to take domains and put it on a second pool since my first one was getting full. I wanted my Docker on one (To host an ARK game server) and Domains on my second pool to try out a Win 11 VM. I started to attempt moving things around ignorantly and ended up how I am now.

 

I am going to move some files with Dynamix now and delete the duplicates now to see what happens. WIsh me luck!

 

 

4 minutes ago, Anthadeas said:

I am going to move some files with Dynamix now and delete the duplicates now to see what happens

Probably simpler than reconfiguring the shares multiple times to get things moved to the array and then back to whichever pool.

 

Remember nothing can move or delete open files.

Hey trurl,

 

So i moved some files around but couldn't get any configuration to actually work. So I went ahead and moved all of the files back to the original SSD and disabled the pool of the other one. My dockers are all running but none of them work still. I just continue to get the attached error. I am not certain what to do now. Should I uninstall all of my dockers and reinstall them with the same settings?

 

server-diagnostics-20220423-2129.zip

https://wiki.unraid.net/Manual/Docker_Management#Re-Create_the_Docker_image_file

 

https://wiki.unraid.net/Manual/Docker_Management#Re-Installing_Docker_Applications