Hi. Need help from the experts please, i don't know if i lost data or not. array not starting. I am at a loss, just recently powered my decade old unraid server.

Thanks.

mckanor-diagnostics-20250708-0830.zip

Not really an expert, but your logs have a lot of errors for:

Disk sdm (WDC_WD20EARS-00MVWB0_WD-WMAZA47938) throwing a lot of errors usually related to power/spinup issues or a controller/cable problem.

a ZFS disk/pool with many errors not mounting, this is likely the main culprit here, appdata,system and domains are looking for "cache" but your only cache disk is showing as a new device.

Nvidia driver needs to be changed to legacy mode as your GPU isn't supported in the new drivers.

One of your "disks" is a hardware raid array of some flavour, which isn't an error but i wouldn't recommend it.

I'm sure someone will help troubleshoot this for you.

Edited July 8, 2025Jul 8 by Faceman

Thanks for the insight bro.

• re the disk , the specific error is "Current pending sector",

• re the ZFS disk/pool, yes, i messed up that cache drive, how can i undo this?

• re the NVIDIA driver, the compatible is still on the way, i will replace this current with that one which is compatible with the latest UNRAID version

• re the "raid" in unraid, yes i have a SAS in this system, thanks for highlighting that, i will be repurposing those SAS drives

However, what would be the easiest path to a solution to get this server up again, it was working a few days only after i did an unclean shutdown, that started it all?

Disk1 appears to be failing, if parity is valid, you can try disabling it and see if the array starts fine, but keep the old disk intact for now.

3 minutes ago, JorgeB said:

Disk1 appears to be failing, if parity is valid, you can try disabling it and see if the array starts fine, but keep the old disk intact for now.

hi. you mean set it to "no device"?

Yep

9 minutes ago, JorgeB said:

Yep

Hi. I had made some HW adjustments, is it still necessary to disable the "soon" to be problematic drive(s)? Only the parity is saying invalid, and it said earlier, if the array started a parity-sync will start. However, it is still trying to start the array.

Something weird, when in the MAIN tab, i pressed the START button to start the array and begin parity-sync, nothing seems to be happening. However, when going to the DASHBOARD tab, the start array "play" button is still enabled and i am able to click that too. But just the same, nothing happening, the array does not start. Here's the latest logs. Thanks for your concern.

mckanor-diagnostics-20250708-1631.zip

Hi. It says "too many errors or disk missing" something like that. I don't know what to do, appreciate your help, i don't mind to rebuild that apps installed, i just want to recover my data, it's 15 years ++ of personal media.

According to that screenshot, parity is not valid.

2 hours ago, JorgeB said:

if parity is valid

Without a valid parity, ddrescue may be the best option to recover as much data as possible from that disk:

https://forums.unraid.net/topic/46802-faq-for-unraid-v6/page/2/#findComment-525075

Just now, JorgeB said:

Without a valid parity, ddrescue may be the best option to recover as much data as possible from that disk:

https://forums.unraid.net/topic/46802-faq-for-unraid-v6/page/2/#findComment-525075

ah ok. how about the rest of my data, on the other disks?

how do i get my array started again and working with all the data on the good disks?

You can do a new config without disk1 (Tools - New config - Keep all assignments - Apply).

Then start the array to begin the parity sync, if any disk fails to mount or there are more read errors, post new diags.

6 minutes ago, JorgeB said:

You can do a new config without disk1 (Tools - New config - Keep all assignments - Apply).

Then start the array to begin the parity sync, if any disk fails to mount or there are more read errors, post new diags.

i checked did that already but with disk 1, and choices are array, pools, right (i choose all) keep same assignments. but it did not work for me. how do i exclude disk 1?

33 minutes ago, JorgeB said:

You can do a new config without disk1 (Tools - New config - Keep all assignments - Apply).

Then start the array to begin the parity sync, if any disk fails to mount or there are more read errors, post new diags.

Thanks, i am on it, will post diags once done. Appreciate it a lot bro.

1 hour ago, JorgeB said:

You can do a new config without disk1 (Tools - New config - Keep all assignments - Apply).

Then start the array to begin the parity sync, if any disk fails to mount or there are more read errors, post new diags.

Hey bro. I got the array back up and running, thanks to all who contributed to the solution. Although i hope the parity-sync will not take years to complete

2 hours ago, JorgeB said:

You can do a new config without disk1 (Tools - New config - Keep all assignments - Apply).

Then start the array to begin the parity sync, if any disk fails to mount or there are more read errors, post new diags.

Thanks bro.

The Parity-Sync has started, it is still reporting read errors on several disks, i will plan to replace those eventually.

I still have a few questions, is there a way to avoid re-installing all apps like Plex and Jellyfin etc. Those that have VMs and Dockers? And can i restore my Cache? I dont know how to recover that, there are appdata folder there, do i also need to finish the Parity-Sync before dealing with restoring or re-installing from scratch the apps? Thanks again brother, the rest of the Unraid support team for helping me.

16 minutes ago, j3ja said:

The Parity-Sync has started, it is still reporting read errors on several disks, i will plan to replace those eventually.

Please post new diags.

17 minutes ago, j3ja said:

Those that have VMs and Dockers? And can i restore my Cache?

You should be able to reimport the cache pool and its data, I assume it was the Samsung SSD? If yes, when you next stop the array, click Add Pool, assign that device, and start the array.

12 hours ago, JorgeB said:

Please post new diags.

You should be able to reimport the cache pool and its data, I assume it was the Samsung SSD? If yes, when you next stop the array, click Add Pool, assign that device, and start the array.

Yes, it's the SAMSUNG. How to do this?

With the array stopped, go to main and click "add pool" to create a new pool with the same name and 1 slot

assign the pool device, leave the filesystem set to auto

start the array to import the pool

1 hour ago, JorgeB said:

With the array stopped, go to main and click "add pool" to create a new pool with the same name and 1 slot

assign the pool device, leave the filesystem set to auto

start the array to import the pool

Thanks, got it

Good day to all, hope you are having a good one.

I couldn't say that for me.

I guess the celebration was premature, as you can see from this ss the disks mounted and the parity-sync was well on its way. Unfortunately, 2 more disks (5&10) were reporting read errors and that caused the server to hang, so needed to shutdown unclean.

The i did the same as before after rebooting, i set those disks to "no device", and did a new config. Then restarted the server. Since this morning, i noticed the ff:

1. no single disk was mounted.

2. no read or write activity on the disks, even as the server is saying "array starting"

3. the server did not hang, i was able to move around and do other stuff

4. i also noticed that if i go run terminal and issue the command "diagnostics", it did nothing and produced no logs or anything

5. i also seen for the first time this error on the monitor (please see ss attached), saying something "panic". I obeserved that prior to this message, running "diagnostics" in the console will produce the diag file, however, if the error already occurred, it will hang if diagnostics is ran.

No idea what to do now, i have attached the logs that i was able to extract from the boot flash drive.

mckanor-diagnostics-20250709-1732.zip

Just now i tried to recreate the event leading up to this error, and i was able to determine that if i don't start the array, this error will not trigger. In the main CPU monitor i can see also that 1 of the cores is getting maxed out. please see ss.

Did you save the diags after the read errors? I had asked for them previously.

The current problem looks like a possible filesystem issue, see if you can get the syslog:

cp /var/log/syslog /boot/syslog.txt

Also, it may be a good idea to run memtest to rule out any obvious RAM issues.