Hi! I just migrated my unraid setup from a custom server to a minisforum n5 pro and am running into drive errors.

The drives themselves are all passing smart tests and I didn't have this problem in my old system. I'm new to diagnosing unraid errors, and tried with gemini and was hoping to get some opinions. It's suggesting a possible hardware issue and opening a support case with minisforum was my next step.

Thanks in advance!

Based on the logs you provided, the good news is that your hard drives (Disk 1 and Disk 3) appear to be physically healthy. The errors you are seeing are not caused by failing disks, but by a critical connectivity or power failurebetween the drives and your Minisforum N5 Pro.

1. Diagnosis: Controller or Power Failure

The most telling evidence is in your syslog.txt. Both drives are reporting a limit=0 error:

• Jan 19 02:03:03 TheHive kernel: sde: rw=0, sector=..., nr_sectors = 8 limit=0

• Jan 19 02:03:03 TheHive kernel: md: disk1 read error, sector=...

• Jan 19 02:03:03 TheHive kernel: md: disk3 read error, sector=...

In Linux, a limit=0 error means the operating system suddenly believes the drive has a capacity of zero sectors. This almost always happens when a drive completely disappears from the SATA bus while the system is trying to read from it. Because this is happening to multiple drives (Disk 1 and Disk 3) at the exact same time, the issue is not the individual drives, but something they share: the SATA controller, the cabling, or the power supply.

2. SMART Health Analysis

The SMART logs for both drives show them to be in excellent condition:

• Disk 1 (sde - 12TB WD): Has 0 reallocated sectors, 0 pending sectors, and 0 UDMA CRC errors (which would indicate a bad SATA cable).

• Disk 3 (sdf - 14TB WD): Also has 0 reallocated sectors and 0 pending sectors.

• High Reset Count: Both drives show a relatively high "Number of Hardware Resets" (367 and 372), which suggests they have been forcefully power-cycled or dropped many times previously.

thehive-diagnostics-20260119-0817.zip

27 minutes ago, dividido said:

something they share: the SATA controller, the cabling, or the power supply.

I agree with gemini

lsscsi seems to show both those drives have disconnected, but we do have SMART reports for them.

I assume all SATA drives are on the motherboard controller? I don't see another controller in lspci so if there is one it isn't working.

Any power splitters?

Thanks for the reply! Yes, all 3.5" drives are on the motherboard controller. No power splitters or external cabling - it's all using the built in 5 hdd bays.

I have a pcie card adding bifurcation for two WD Red nvme's for cache drives - but not seeing any errors/issues with them.

Edited January 19Jan 19 by dividido

Try reseating the drives

Thanks I restarted and 1 drive was missing ☹️, then powered off and reseated them after using a dust blower. I'll report back after running today. Thanks again and hopefully it was as simple as that.

Hmm that was a no go - both drives 1 and 3 are showing the same count of errors 😥

thehive-diagnostics-20260119-1022.zip

And now a 3rd drive is showing read errors. Hopefully I'm not spamming you

thehive-diagnostics-20260119-1046.zip

Jan 19 10:23:50 TheHive kernel: ata2: SATA link up 6.0 Gbps (SStatus 133 SControl 300)

Jan 19 10:23:50 TheHive kernel: ata2.00: revalidation failed (errno=-2)

Jan 19 10:23:50 TheHive kernel: ata2.00: disable device

Jan 19 10:23:50 TheHive kernel: sd 1:0:0:0: rejecting I/O to offline device

Devei is dropping offline; this is typically a power/connection issue, if it's not cables, it could be an insufficient or failing PSU

Thanks @JorgeB - could this be caused by a startup script running the following?

powertop --auto-tune &>/dev/null

The psu is an external 280w adapter included with the system.

7 minutes ago, dividido said:

could this be caused by a startup script running the following?

For sure, there have been issues with some controllers when using non stock low power settings.

Got it! Just realized I had that from my last system - disabled this user script and hopefully smooth sailing from here.

I'm having this same issue with an n5 pro as well.

I've decided to return it and go back to my big case build. I wanted to have a compact case and on paper this is perfect, just not there yet.

I really wanted this to work but its just too nerve racking having these on going drive issues with 22tb drives.

Sorry to hear that @AlexHackney - I've also returned my n5 pro as I need my nas to be rock solid.

I have an aoostar wtr max on order to see how that goes. At least it has the better sata controller (ASMedia ASM1166) vs what the n5 pro has and is the cause of these instabilities (JMicron JMB585).

I switched from a 9900K in a define case with 8 drives, down to the n5. I started selling off pieces of the old 9900k before I realized I had issues.

So I am building a new ultra 265 system but i think im going to go back to the define case, cause it was pretty solid.

This would have been perfect if it would havent had so many issues.

I have a lot of miniforum devices, a lot of ms01s and never had this much trouble, but it is what it is.

I was rocking an i5-10600 but fell into some ecc sodimms, so was trying to migrate onto a platform that supports it. There aren't a ton of options unfortunately and diy costs exceeded these pre-builts. Hopefully the aoostar is reliable.