HI All,

My parity device was/is disabled so I followed the steps that have been recommended:

1. Stopped the server.

2. Unassigned the parity drive.

3. Start up the server.

4. Stop server again.

5. Reassign the parity drive.

6. Start up the server.

Maintenance mode was NOT used at all in the above.

The rebuild begun and was done last night however, the parity drive is still disabled. Attached are the diagnostics, any help would be appreciated!

datavault972-diagnostics-20251109-0843.zip

Whatever the reason was for the disk to be disabled again sadly isn't visible in the logs because the server was rebooted and the log of the previous run is incomplete because it's filled with overheating messages from your HBA.

Nov 8 05:29:54 DataVault972 kernel: mpt2sas_cm1: Temperature Threshold flags 0 exceeded for Sensor: 0 !!!

Nov 8 05:29:54 DataVault972 kernel: mpt2sas_cm1: Current Temp In Celsius: 70

Nov 8 05:30:24 DataVault972 kernel: mpt2sas_cm1: Temperature Threshold flags 0 exceeded for Sensor: 0 !!!

Nov 8 05:30:24 DataVault972 kernel: mpt2sas_cm1: Current Temp In Celsius: 70

Nov 8 05:30:54 DataVault972 kernel: mpt2sas_cm1: Temperature Threshold flags 0 exceeded for Sensor: 0 !!!

Nov 8 05:30:54 DataVault972 kernel: mpt2sas_cm1: Current Temp In Celsius: 70

Would start by addressing cooling.

Damn! So should I try and redo the steps that I did and then get the diagnostics?

I am prepping to put my server in a proper cabinet with better cooling, for now I will redo my fans.

Try improving the HAB cooling, then try again to resync parity and post new diags if it still fails.

40 minutes ago, JorgeB said:

Try improving the HAB cooling, then try again to resync parity and post new diags if it still fails.

I have removed the top panel from my PC server and increased the speed of the fans that are on the HBA. Hopefully that helps. Is there anyway to track the HBA temps in unraid? The fans are PWN so it would be cool if they PC/software could control the speed as needed

19 hours ago, Tuntira said:

Is there anyway to track the HBA temps in unraid?

Not that I'm aware of in the GUI, seem to remember something with the CLI.

On 11/9/2025 at 2:44 PM, JorgeB said:

Try improving the HAB cooling, then try again to resync parity and post new diags if it still fails.

I worked on my cooling and also redid the syncing. It just finished and attached are the diags

datavault972-diagnostics-20251110-2140.zip

HBA is still reporting overheating, though 50C seems reasonable, the disk errors look more like a power/connection problem.

12 hours ago, JorgeB said:

HBA is still reporting overheating, though 50C seems reasonable, the disk errors look more like a power/connection problem.

I'm thinking of buying new SATA cables, although I am not sure if that will solve the problem.

You mentioned power/connection problem, do you think that it could also be the PSU. I and using a Fractal 550W PSU (https://www.fractal-design.com/products/power-supplies/ion/ion-2-platinum-560w/black/). Is ther anything in the diags that would point to the PSU?

11 hours ago, Tuntira said:

Is ther anything in the diags that would point to the PSU?

Not directly, but a power issue could be the connectors or the PSU.

hi @JorgeB could you assist me with my Parity drive, for some reason it is showing as disabled and has been so for a while. Everything else has been working fine with no dropouts from anyother drive for the past 10 days.
I did replace the cable for this drive and I am planning on replacing the other cables too. I bought a new PSU but have not installed it yet. I was hoping to test the cables first and if I find that the PSU is NOT the problem, then I can return it.
I have attached diags

datavault972-diagnostics-20251130-1258.zip

The syslog is being spammed with HBA temperature-related errors. It doesn't show the issue with parity, but since there's no SMART looks like it dropped offline. Try replacing or swapping both cables with a different disk and retest.

Alrighty,
I have fixed the cooling again (Fans weren't connected).

I swapped the cables completely, hoping that something shows up

datavault972-diagnostics-20251130-1626.zip

It won't show how the disk got disabled, but it does show the SMART report, and that looks mostly fine, so my recommendation from before remains.

You can also run an extended test on disk just to make sure it's OK, since there are a few reallocated sectors.

Hi @JorgeB I did some testing and it seems like Drive 8 has failed. When I plug it in and run the server, it does not spin up at all. I have felt it and listened to it.

I think that it may have died. If that is the case. How can I start up my array as, with both the parity drive and drive 8 down, I am not able to start it up.

OK I did some further testing, I plugged Disk 8 into an external enclosure and it failed to spin up and even be recognized by Unraid. Whereas, when I plugged in the parity drive, it was picked up and spun up.

This confirms that the drive IS dead.

Apologies, I know that this is a very similar message to what I sent an hour ago, but it is confirmation.

My problem is now, how can I replace the DISK 8 and then get the parity to rebuild itself

You can do a new config (Tools - New config) to be able to start the array with the remaining drives, but note that any data that was on disk8 will be lost.

22 minutes ago, JorgeB said:

You can do a new config (Tools - New config) to be able to start the array with the remaining drives, but note that any data that was on disk8 will be lost.

Which one of the above must I choose?

Keep all

Nearly DOne! Crossing fingers that everything will sort it self out when the parity is done

1 hour ago, Tuntira said:

Crossing fingers that everything will sort it self out when the parity is done

There are a lot of disk errors shown, there are still issues, and also unmountable disks. Post new diags, please.

So the parity check finished and then I formatted the 3 new drives and then mounted them. Everything is looking good.

Is there anything that I need to be worried about from the previous post? Disk4 is a drive that was previously in the array and Disks 8, 11 and 12 are the new ones.

I added diagnostics for clarity.

datavault972-diagnostics-20251216-1103.zip

3 hours ago, Tuntira said:

Is there anything that I need to be worried about

Yes, all the disk errors are happening with multiple disks:

Dec 16 09:27:13 DataVault972 kernel: md: disk2 read error, sector=1845846096

Dec 16 09:27:13 DataVault972 kernel: md: disk9 read error, sector=1845846096

Dec 16 09:27:13 DataVault972 kernel: md: disk4 read error, sector=1845846096

Dec 16 09:27:13 DataVault972 kernel: md: disk3 read error, sector=1845846096

Dec 16 09:27:13 DataVault972 kernel: md: disk4 read error, sector=19387713000

Dec 16 09:27:13 DataVault972 kernel: md: disk3 read error, sector=19387713000

Dec 16 09:27:13 DataVault972 kernel: md: disk2 read error, sector=19387713000

Looks more like a power/connection issue, any power splitter in use? It could also be bad or weak PSU

Below is a drawing of the power connections in my system.

Based on the info that you shared, the power/connection issues are happening on one branch.

PA = Parity drive

Numbers = the drive number in Unraid. (1 = Disk1, 2 - Disk2 etc)

Edited December 16, 2025Dec 16 by Tuntira

This is a terrible idea; avoid using splitters. And if you do, don't split one SATA port into more than two.