Had a Truenas server built from some old parts previously. Decided to make the change to Unraid. Bought a whole new setup with new HHDs to take my server to the “next level”. That is until one of my new HDDS (Disk 1) decided to become a potato.

 

There are a few issues that I have now realized with my build that I will get addressed but this is my setup (Prior to disk failure)

 

CPU: Ryzen 9 7900X

Motherboard: PRIME B650M-A AX II

(2x16GB) x 2 (64GB) – DDR5-6000

Parity

•          12 TB HHD

Array Devices

•       12 TB HHD (This is the Disk that Failed)

•       12 TB HHD

•       12 TB HHD

•       8  TB HHD

•       1   TB HHD

Cache Pools

•          6 TB HHD (Need to change this, found that out early on..)

•          250 GB SSD (Old SSD that I had been using for certain Dockers)

 

Once I was confident enough to start transferring a large amount of data, Disk 1 failed during a transfer. Spent countless hours diagnosing it as a novice, but could not figure it out. Below are some lines from syslog.txt while I still had the drive connected.

 

ata1.00: Read log 0x10 page 0x00 failed, Emask 0x1

Jan 16 21:51:54 Hometower kernel: ata1: failed to read log page 10h (errno=-5)

Jan 16 21:51:54 Hometower kernel: ata1.00: exception Emask 0x1 SAct 0x40000 SErr 0x0 action 0x6

Jan 16 21:51:54 Hometower kernel: ata1.00: irq_stat 0x40000001

Jan 16 21:51:54 Hometower kernel: ata1.00: failed command: READ FPDMA QUEUED

Jan 16 21:51:54 Hometower kernel: ata1.00: cmd 60/08:90:00:00:00/00:00:00:00:00/40 tag 18 ncq dma 4096 in

Jan 16 21:51:54 Hometower kernel:         res 00/00:00:00:00:00/00:00:00:00:00/00 Emask 0x3 (HSM violation)

 

Frustrated, I decided to remove it from the array and replace it with a smaller old 1 TB HHD that I had. (I believe) This is not ideal, but I was following this guide, specifically under -> Replacing Disks -> Replacing Failed/Disabled Disks -> Normal Replacement.  “This is a normal case of replacing a failed drive where the replacement drive is not larger than your current parity drive(s)”

 

Here is where I think I may of messed up.. I followed another guide (Will have to dig further in search history to find link) that mentioned to use “New Config”. Did this and checked “ALL” for “Preserve current assignments”.

 

Mounted my old 1 TB HHD and started array and did not see any data. I quickly shut it down and could not find anything on forums to help with this situation.

 

Any ideas on how to recover?

 

Diagnostics are attached.

 

 

hometower-diagnostics-20250118-0014.zip

You can't replace a drive with a smaller drive. 

 

The only way to get it to allow that assignment is New Config, but that will only rebuild parity so no data rebuild possible. 

 

Possibly there was nothing wrong with the original disk, and it may still have its data. 

 

On mobile now can't look at Diagnostics yet. I will take a look in the morning. Probably someone else will advise before I can. 

 

Don't do anything else without further advice. 

Reconnect old disk1, leave it unassigned, and post new diags.

Very unlikely you have deleted any data. One of the best things about Unraid is that each array data disk is a separate filesystem that can be read all by itself on any Linux. So even if the problem disk is dead (unlikely), the data on all the other disks is unaffected.

 

Probably even the original disk still has its data.

 

Did you start the array after New Config?

 

You have debug mode enabled in Unassigned Devices

Quote

 Use only when directed for support because the messages can flood the log.

and it is indeed flooding the log. Turn that off before posting new diagnostics.

 

 

 

3 hours ago, JorgeB said:

Reconnect old disk1, leave it unassigned, and post new diags.

 

On 1/18/2025 at 8:02 AM, trurl said:

You have debug mode enabled in Unassigned Devices

 

Okay, I have turned that off. Restarted system and attaching new diags. (This does not have my old Disk1 included).hometower-diagnostics-20250119-1102 - Debug Mode Off.zip

 

On 1/18/2025 at 4:10 AM, JorgeB said:

Reconnect old disk1, leave it unassigned, and post new diags.

 

Okay, new diags attached with my old Disk1 attached. hometower-diagnostics-20250119-1102 - Debug Mode Off - Old Disk1 Included.zip

One thing to note, I am unable to detect this disk in bios. I wasn't able to previously after it became disabled. I have tried switching SATA ports / cables, & power. No change.

 

 

On 1/18/2025 at 8:02 AM, trurl said:

Did you start the array after New Config?

 

Yes... That's when I noticed I had an issue

1 hour ago, viper28 said:

unable to detect this disk in bios

Unraid doesn't see it either. Can you try plugging it in some other way? USB enclosure, for example?

 

1 hour ago, viper28 said:

Yes... That's when I noticed I had an issue

Normally, you would let it rebuild parity after New Config, but there is a checkbox to tell it parity is already valid.

 

If you didn't let it rebuild parity there might be some chance that we could get the parity that existed before the New Config to emulate the missing disk and let us rebuild it. But if you did rebuild parity then there is nothing that can be done except hope the original disk can be read somehow.

Just now, trurl said:

plugging it in some other way? USB enclosure, for example?

Even better if you could get it detected by plugging it into another port, cable, etc.

Just now, trurl said:

Unraid doesn't see it either. Can you try pluggin it in some other way? USB enclosure, for example?

Yes, I can plug in USB, new diags attached - hometower-diagnostics-20250119-1315 - Disk1 Attached Via USB.zip

 

Looks like Unraid detects it -

 

 

 

Lots of errors on disk log- 

 

 

I checked "Parity is Valid" before starting the array that one time, so hopefully nothing was done... 

1 minute ago, viper28 said:

Looks like Unraid detects it

Not getting a good SMART report for it though, and Unassigned Devices isn't offering to mount it. Don't know if that is because of the disk or because of the USB.

 

Did you try another SATA port and/or SATA cable and/or power cable?

 

3 minutes ago, viper28 said:

I checked "Parity is Valid" before starting the array that one time, so hopefully nothing was done... 

Do you know if anything has been written to the array since then?

Just now, trurl said:

Did you try another SATA port and/or SATA cable and/or power cable?

Yes, I have tried a bunch previously.

1 minute ago, trurl said:

Do you know if anything has been written to the array since then?

Once I started the array I immediately stopped. However, I am sure my dockers were on so I assume so...

On 1/18/2025 at 12:16 AM, viper28 said:

Frustrated, I decided to remove it from the array and replace it with a smaller old 1 TB HHD that I had. (I believe) This is not ideal, but I was following this guide, specifically under -> Replacing Disks -> Replacing Failed/Disabled Disks -> Normal Replacement.  “This is a normal case of replacing a failed drive where the replacement drive is not larger than your current parity drive(s)”

https://docs.unraid.net/unraid-os/manual/storage-management/#normal-replacement

Quote

You must replace a failed disk with a disk that is as big or bigger than the original and not bigger than the smallest parity disk.

 

On 1/18/2025 at 12:16 AM, viper28 said:

Here is where I think I may of messed up.. I followed another guide (Will have to dig further in search history to find link) that mentioned to use “New Config”

https://docs.unraid.net/unraid-os/manual/storage-management/#reset-the-array-configuration

The warning screenshot at that link shows what you should have seen when you did this:

Quote

DO NOT USE THIS UTILITY THINKING IT WILL REBUILD A FAILED DRIVE - it will have the opposite effect of making it impossible to rebuild an existing failed drive - you have been warned!

 

6 minutes ago, trurl said:

and Unassigned Devices isn't offering to mount it. Don't know if that is because of the disk or because of the USB.

I did have the option to mount it under the "Array Devices" I clicked on it, but then nothing changed, still showed up as, "Unassigned" for Disk 1

 

However, the Device ID changed under unassigned disks

 

 

2 minutes ago, trurl said:

DO NOT USE THIS UTILITY THINKING IT WILL REBUILD A FAILED DRIVE

Yeah, I misread that entirely. Embarrassed to share what I thought it meant.... 

 

Do I need to start looking into disk recovery options on another machine and transfer data? (If that is even possible) Then start a new Unraid from scratch?

 

 

17 minutes ago, viper28 said:

Once I started the array I immediately stopped. However, I am sure my dockers were on so I assume so...

All of the diagnostics you have posted so far are with the array not started, or possibly started in Maintenance mode. Nothing could read or write the array if you never started it in Normal mode. Also, your docker and VM related shares (appdata, domain, system) are configured to not be on the array. Can't tell if they do have files on the array since the array was not started in Normal mode.

 

If you did have it started in Normal mode, then even if those shares were not on the array, possibly some of your containers might have been configured to write to other shares that are on the array.

 

But if the array wasn't started long it might be possible to force Unraid to rebuild the missing disk. We can try that, and even if it isn't a perfect rebuild, maybe we can still get some or most files from the rebuild, possibly after repairing the filesystem.

 

Another approach would be to see if we can repair the filesystem on the Unassigned Disk. Currently UD thinks it can't be mounted so it is offering to format it instead. But maybe repair would make it mountable and so we could recover some or most of its data if the disk is still working well enough.

 

Do you have a spare disk (at least as large as the original disk but no larger than parity) that you can use to try the rebuild?

1 minute ago, trurl said:

Do you have a spare disk (at least as large as the original disk but no larger than parity) that you can use to try the rebuild?

No, both Parity and Disk 1 are 12 TB. I do have (2) other 12TB disks that are installed on the array. Assuming with little to no data on the latter...

Just now, viper28 said:

I do have (2) other 12TB disks that are installed on the array. Assuming with little to no data on the latter...

All of the existing disks are needed to emulate the missing disk. Doesn't matter at all whether they are empty or not. You must use a different disk.

 

https://docs.unraid.net/unraid-os/overview/nas/#parity-protected-array

On 1/18/2025 at 12:16 AM, viper28 said:

Had a Truenas server

Parity basically works the same way in any system where it is used. Parity is just an extra bit that allows a missing bit to be calculated from all the other bits.

 

15 minutes ago, trurl said:

All of the existing disks are needed to emulate the missing disk. Doesn't matter at all whether they are empty or not. You must use a different disk.

I am unable to purchase an internal 12TB HHD locally today. I will go ahead and purchase a new 12TB drive now. Once that comes in, how should I approach the rebuild?

4 minutes ago, viper28 said:

Once that comes in, how should I approach the rebuild?

Shutdown until you are ready to do the rebuild.

 

Basically we will New Config/Trust Parity with the new disk in place, then disable the new disk so Unraid can rebuild it.

 

Let us know when you are ready to proceed. It is very important that you follow our instructions exactly.

2 minutes ago, trurl said:

Let us know when you are ready to proceed. It is very important that you follow our instructions exactly.

Understood.

 

Disk should be here tomorrow.

 

Thank you for your help thus far.

On 1/19/2025 at 2:31 PM, trurl said:

Let us know when you are ready to proceed.

New disk is in with the system off, ready for your advice when you have the availability.

Post new diagnostics

3 minutes ago, trurl said:

Post new diagnostics

New diagnostics attached - hometower-diagnostics-20250120-1715.zip