trurl

And you should post diagnostics while you can. Also setup Syslog Server so syslog will be saved so you can get it after crash: https://forums.unraid.net/topic/46802-faq-for-unraid-v6/?do=findComment&comment=781601

Have you done memtest?

Might as well put flash in your PC while you are checking those connections, and let your PC do checkdisk on it. While there, make a backup of it on your PC. Answer the questions, check connections, checkdisk on flash and make backup, and do memtest. Then post another diagnostic and wait for a reply before proceeding with filesystem check.

As Squid alluded to flash drive is probably at the root of some of your problems. What kind of flash drive are you using? Is it plugged in to a USB2 port (recommended)? Your diagnostics says parity is invalid. Have you ever completed the initial parity sync? Looks like the SMART warning on disk2 is just CRC errors. So the disk is probably OK but has had connection issues. You can click on the warning to acknowledge it and it will warn again if it increases. Possibly the connection issue is the reason for the unmountable filesystem though. Shutdown, check all connections for all disks, power and SATA, both ends, including power splitters. When you reboot make sure you are using a USB2 port, and instead of booting into Unraid, do memtest just to get that out of the way. Do you have a backup of flash?

Are you trying to access by IP address or by server name?

Your question makes me wonder if you understand that Unraid IS NOT RAID. Parity has no data, and each data disk is an independent filesytem that can be read by itself on any linux.

Don't assign parity and nothing will be written to any drive. Or assign parity and only parity will be written. Nothing will happen to your data (that hasn't already happened) unless you accidentally assign any data disk to any parity slot or agree to format anything. You will automatically have all your user shares, but they will all have default settings. The user shares are simply the top level folders on all the disks and the user share names are those folder names. Whether or not you have created a user share, if a top level folder exists on cache or array it is a user share.

Note that the only time a disk needs to be clear is when adding it to a new data slot. A clear disk isn't needed when replacing a disk, or when adding a parity disk, since that disk is going to be completely overwritten from the parity calculation anyway. And as mentioned, if you don't preclear a disk, Unraid will clear it if it is added to a new data slot.

There is also the ability to just select a template from flash with the Select a template dropdown on the Add Container page. That has always been there and is there whether or not you even have CA installed.

The obvious difference between the two speeds is gb ethernet vs wifi. What is the question?

I haven't run any VMs on my server in years since docker does everything I need, so I will let someone else address that question. As for upgrading, you should always have a backup of flash. I don't actually know if that old version does upgrades from the webUI or not, but you can always just do a new install and copy the config folder from your flash backup. But there are some other things you need to take care of before upgrading from that old version. See here: https://forums.unraid.net/topic/66327-unraid-os-version-650-stable-release-update-notes/

One of the choices with New Config is to keep all the assignments. All that means is it won't forget any, but you can still change them including unassigning any or moving them to different slots. So keeping only parity isn't really necessary to allow you to change the others. But if you have done that then every thing should still be fine. There are only two thing that can cause loss of data. One is assigning any data disk to any parity slot, since that would result in overwriting the data with parity. The other is formatting any disk that you know has data on it. Just in case something goes wrong with a connection or something, don't agree to format anything.

Unraid just gets the data for each user share completely based on which top level folders exist on each disk. The user shares ARE the top level folders with the same name as the user share. That is how it works for reading the user shares. For writing the user shares the settings for each user share are considered in addition to which disks contain them, and the user share settings take precedence, so for example if you exclude a specific disk, that excludes it for writing, but for reading any disk with the folder is included.

You talk about "turning Unraid off" several times. What exactly do you mean by that? You must always shutdown from the webUI. Probably you have just broken user shares with a corrupt cache or something but most of the data is OK. Start the array again and go to Tools - Diagnostics and attach the complete diagnostics zip file to your NEXT post.

Why are you running a nearly 3 year old version of Unraid? Can't you just access the drive as an Unassigned Device on the host?

The way I usually put it is: Check connections, SATA and power, both ends, including any power splitters. Could also be a bad cable, SATA or power. Could be a bad port. I usually defer to more knowledgeable people on controller issues that involve multiple drives. Could be bad drive but the SMART we got earlier seemed OK. It's not always clear to me how to interpret those in the case of SSDs of various manufacture though.

For anybody stressing out, the best way to keep calm is to always have another copy of anything important and irreplaceable. Parity, even dual parity, is no substitute for a backup plan.

There is some benefit to having appdata cache-prefer, since it is possible that some applications might have appdata grow to the point of needing to overflow to the array. But just be aware that if it does overflow to the array, there is no way for Mover to move it back to cache with the docker running since Mover can't move open files. I just always make sure that cache has plenty of room. There is no benefit to making system share cache-prefer since that shouldn't grow anyway. I think the reason those shares default to cache-prefer is because some people might install cache after setting them up, but it doesn't really help since nobody knows that they have to stop Docker and VM services to get them moved. More important than all that though, is making sure that cache has Minimum Free set appropriately in Settings - Global Share Settings. In order for anything to overflow, Unraid must decide that cache doesn't have enough room. Unraid has no way to know how large a file will become when it begins to write it. If cache has less than Minimum, Unraid will choose another disk before it begins to write the file. You must set Minimum Free to larger than the largest file you expect to write to cache. Each User Share also has a Minimum Free setting that works in a similar manner. If Unraid sees a particular disk has less than Minimum, it will choose another disk, unless Include/Exclude or Split Level prevent it from doing so. You can always see which disks each of your User Shares is using by going to Shares - User Shares and clicking Compute... for the share, or just use the Compute All button at the bottom of that page.

The discussion has moved to the docker support thread here: If you wish to comment further take it there. I am locking this one.

By "my NAS" do you mean a different machine than you are running the docker on? NVM I just realized you're the same person on that other thread. You should have posted a link back to it in your post to this thread. Here it is for everyone else to see: Now that we are over here, I am going to lock that thread after linking it here to this one so we can all work together.

No SMART for cache again.

I know you said you tried all USB ports, and I looked up that mobo. It does have USB2. USB2 is more reliable for booting. Just thought I would mention that since it wasn't actually said on this thread. What kind of flash drives were you trying? Maybe you didn't try all of the different things you mentioned on those USB2 ports. Maybe try them again but only USB2.

user0 is just the user shares excluding cache. Not sure why you would want to do that since the best place for appdata is the cache drive. For Unassigned Devices, are you referring to shares it has mounted from other computers? Just the base UD plugin is enough for that. I generally recommend keeping everything up-to-date.

Not quite clear for me. Is there something confusing about the way I said it?

There should only be a single partition on the Unraid boot flash. If you had a good backup you could just make a new install and copy the config folder from your backup. The config folder is where all of your settings are. That old backup won't have any settings you made since then. Most importantly, if you made any changes to disk assignments, those would be wrong. One possible scenario is using a backup which specified a disk assigned as parity that you had since reused as data. If you used that Unraid would begin to overwrite that data disk with parity. Obviously having a flash backup on a system which can only be accessed after you have booted from flash is not the best plan. I use an NTFS Automounted Unassigned Device for my scheduled flash backup, that way I can just plug it into Windows if I need it. Another method would be an Automounted share on another computer in Unassigned Devices. Neither is a perfect solution though since if they aren't mounted for some reason the path to those would be in RAM instead. Do you have any diagnostics, syslogs, or screenshots we could get your disk assignments from? Do you have a copy of your license key? If we could get that far you and get the array started you could retrieve your backup.