trurl

Where did you find this "guide"? Sounds like it led you down the wrong path. New Config isn't typically part of drive replacement, assuming they actually needed replacing. Possibly it was simply a case of the much more common bad connection. And even if you do New Config, it shouldn't forget your disk assignments unless you tell it to. Please consider asking for advice before trying to fix these sorts of problems in the future. Are you sure you didn't actually lose data winging it like this? Also, your system share has files on the array, and your docker image is twice as large as recommended.

Settings - Display Settings - Temperature unit:

Sounds like maybe a hardware problem could be the root of your other problem. Is the USB drive going to sleep or something like that?

Ok, that's the first time you said you actually did that. Since this is a new mapping, I am a little unclear what you mean by this: Did you add them to the library and let it scan for the new media?

And that would seem to indicate that the plex library (appdata) is working, but it can't find the media. If you don't use slave mode, then the docker service won't be able to see the unassigned device until you restart the service.

If you don't use slave mode then you have to restart the docker service (not the container, the actual docker service) after the Unassigned Device is mounted.

Also, your Unassigned Device has filesystem corruption. Are your dockers using it?

And why do you have 100G allocated to docker image? 20G should be more than enough, and when a user makes it as large as you have then I have to wonder if they have been filling it up. Have you?

Have you done memtest?

In fact, I wonder if it is even on a network, since you couldn't reach it from another computer. Is it showing an IP address in the upper right corner of the webUI?

Can your server reach the internet?

Are you using the browser you get on the server itself when booted in GUI mode? Or do you mean using a browser on another computer on your network?

Of course, you just have to map them in the container for each of your dockers that you want to have access.

If it works in Windows why not update its firmware there?

I doubt it since this functionality is planned for core Unraid soon©

LAN makes more sense to me. Why go to a lot of extra trouble to save a little bit of time for a one time operation?

That screenshot of your Krusader configuration may not be showing us the whole picture. Post the docker run command for Krusader. https://forums.unraid.net/topic/57181-real-docker-faq/#comment-564345

It also occurs to me that there might be some misunderstanding about Unassigned Devices shares, and Unraid User Shares. Unassigned Devices plugin lets you share, on the network, devices that are not part of Unraid cache or array. These will show up in Windows, for example, next to your Unraid User Shares. But they are not part of Unraid User Shares. If you move one of your user shares completely over to an Unassigned Device, it is no longer an Unraid User Share.

And if you did actually move (not simply copy) those user shares off of cache or array drives, and onto an Unassigned Device, then you have effectively removed those user shares. They will probably still show up, but they are not really there, and they would disappear when you restarted the array or rebooted. EXCEPT, probably the system share would be recreated with a new docker image since docker service is probably enabled. So, your description is pretty confusing. I hope your system isn't actually in that much confusion.

Your system, domain, and appdata folders (not uppercase) are Unraid user shares. Since I don't know the details of how you may have changed the Krusader mappings, I don't know what you would see within the Krusader application itself. And here is where I am unclear what specifically you mean by "Shares". Where did they "appear". What exactly are you looking at when you say that. Windows? Krusader? The Unraid webUI? To elaborate on this, any top level folder on cache or array is automatically a user share with the same name as the folder. If Unraid now has user shares named "UNASSIGNED" and "NVMe", you must have accidentally created them as top level folders on cache or array, probably due to the way you moved or copied.

If you had asked for advice sooner, this is what I would have recommended. It's really just a case of what the hardware is capable of. There is simply no way to move stuff from the faster cache to the slower parity array as fast as you can write it to the faster cache.

Try testing individual modules in different slots to see if it is a specific RAM module or a specific slot giving the errors.

Note that the default btrfs raid1 gives a mirror, so you would only get the capacity of the smaller drive. There are other possible configurations: https://forums.unraid.net/topic/46802-faq-for-unraid-v6/#comment-480421

Actually your setup doesn't look that bad. Better than a lot of people. Many people setup the dockers, etc. before installing cache and then I have to work with them to get things that belong on cache moved or reinstalled to cache. But looks like you are good with that, except as you say, your plex library. And your cache has plenty of space for that. I would just copy the plex library to its usual appdata location on cache and change the docker to use it there instead of the UD where you have it now. Then your problems with appdata backup are solved.

You can create a cache pool of SSDs and have redundancy that way. You would have to also have at least one data drive in the array in order to start, but you could conceivably work with just the cache pool. https://forums.unraid.net/topic/46802-faq-for-unraid-v6/#comment-480421