barryfjohnson

This is the URL to the video I've followed:

I wasn't intending to go that route, but by accident - I did. I went ahead and Started the array last night, it's still doing the parity rebuild. Everything seems to be going as I would expect. I expect I have a few more hours to go. Fingers crossed, but I do feel good about what's taking place.

I was following Space Invader's video on Shrinking an Unraid Array. I have removed a disk before using his video (2 weeks ago) and had no problems. Today I was following the same instructions, got to the point where I shutdown the array and removed the disk, then rebooted the array. Everything looked as I was expecting from the video and my earlier disk removal. The next thing he has you go to the Tools menu, pick New Config, under 'Preserve current assignments' you pick 'Pool Slots'. When I went back to the Main tab, my 2 Cache Disks looked right, but my 2 Parity Disks and my 4 Data Disks were all showing "Unassigned" ---> I know that's bad. I went to each of them and did the drop down and chose the disk from the choices given for that Disk location. I had documented everything about what disk was assigned to each Dev in unRaid. There is a RED notice next to each of the 2 Parity disks that says "All existing data on this device will be OVERWRITTEN when array is Started". I'm afraid to go any further. I don't know if when I start the array it will rebuild the 2 Parity disks as they should be OR if my Data is about to go away... I seriously need some experienced help on what my next move should be. Thank you.

Quick update... I rebuilt Unraid from my backup file on a new motherboard, new CPU and new memory. Problem went away. Server shuts down (powers off) when I tell Unraid to do a Shutdown.

In the thread there was a reference to use the command "/sbin/poweroff" to see what would happen shutting the OS down that way. [note: I am NOT a Linux person, barely a MS-DOS guy, but I'm not afraid to try commands] I booted the system, but I didn't Start the array from a browser like I normally do. It was at the login prompt on the console "Tower login:". I logged into root and entered the command "/sbin/poweroff". The OS hung on the way down at the same place as it does when I remote in and Start unRaid from a browser, then tell it to Shutdown: ... /boot : successfully unmounted /run : successfully unmounted /sys : ignored /proc : ignored umount: /: not mounted. It didn't power off. Just hung right there at "umount: /: not mounted." as before. Not sure what that tells me. I feel pretty sure the hardware is OK. Something has changed on one of my OS updates I guess. I don't have any VMs, but I do have these 4 Plugins: Community Applications, Dynamix System Information, Unassigned Devices and Unassigned Devices Plus The other thing I couldn't help but notice is how there are big delays at some places on startup and shutdown when I watch the console. That's new. The OS used to boot and shutdown pretty fast.

I won't replace the discs (~ 16Tb) either.

I tried the timer mods that were on the oldest thread. No change. On the shutdown, everything looks normal, but the last line, as mentioned above is "umount: /: not mounted." and the server just sits there and never powers off. My server is running on hardware that's almost 13 years old... So I'm going to not spend time on this power off issue. I'm going to order up new equipment on the Black Friday sales next week and replace everything but my case. Thank you for the reply. I may browse the rest of the thread and see if something else pops up that might solve the issue. My data contains items for work & family that I really don't want to loose, I'm OK with spending a few hundred to get current and refresh the hardware. This is an awesome backup solution. I've had 2 disc failures about 3 years apart. Both times the recovery was as expected when I replaced the discs thanks to Lime's software!

Same issue for my server. Sits at that same state until I turn off the power supply. Always did a graceful shutdown, then a power off in the past. Maybe someone will uncover the issue.