Hello,

today I noticed that the parity check is running and is taking forever. The Speed was very low and I noticed some "funny" noises from one of the HDDs of the Array.

The Errorlog tells me something like this in a cyclic way.

Mar 15 12:38:18 curie kernel: ata1.00: exception Emask 0x10 SAct 0x700000 SErr 0x4090000 action 0xe frozen
Mar 15 12:38:18 curie kernel: ata1.00: irq_stat 0x00400040, connection status changed
Mar 15 12:38:18 curie kernel: ata1: SError: { PHYRdyChg 10B8B DevExch }
Mar 15 12:38:18 curie kernel: ata1.00: failed command: READ FPDMA QUEUED
Mar 15 12:38:18 curie kernel: ata1.00: cmd 60/40:a0:c0:e5:13/05:00:3f:00:00/40 tag 20 ncq dma 688128 in
Mar 15 12:38:18 curie kernel: ata1.00: status: { DRDY }
Mar 15 12:38:18 curie kernel: ata1.00: failed command: READ FPDMA QUEUED
Mar 15 12:38:18 curie kernel: ata1.00: cmd 60/80:a8:00:eb:13/00:00:3f:00:00/40 tag 21 ncq dma 65536 in
Mar 15 12:38:18 curie kernel: ata1.00: status: { DRDY }
Mar 15 12:38:18 curie kernel: ata1.00: failed command: READ FPDMA QUEUED
Mar 15 12:38:18 curie kernel: ata1.00: cmd 60/40:b0:80:e0:13/05:00:3f:00:00/40 tag 22 ncq dma 688128 in
Mar 15 12:38:18 curie kernel: ata1.00: status: { DRDY }
Mar 15 12:38:18 curie kernel: ata1: hard resetting link
Mar 15 12:38:24 curie kernel: ata1: link is slow to respond, please be patient (ready=0)
Mar 15 12:38:25 curie kernel: ata1: SATA link up 1.5 Gbps (SStatus 113 SControl 310)
Mar 15 12:38:25 curie kernel: ata1.00: configured for UDMA/33
Mar 15 12:38:25 curie kernel: ata1: EH complete
Mar 15 12:38:38 curie kernel: ata1: link is slow to respond, please be patient (ready=0)
Mar 15 12:38:39 curie kernel: ata1: SATA link up 1.5 Gbps (SStatus 113 SControl 310)
Mar 15 12:38:39 curie kernel: ata1.00: configured for UDMA/33

I attached the Diagnostics for any help!

I am not sure what's going on. I checked the cable connection of the Drive and replugged it. The Server was off for this purpose.

Any help is appreciated! TIA!

curie-diagnostics-20260315-1240.zip

Looks like a power/connection issue with disks 2 and 3, check/repalce cables, any power splitters in use?

Hey,
The System was running for a long time in this configuration.
I am debating if it could be that the power distribution is subpar.

All HDDS are connected to the N100 Board and its SATA_PWR1 port, which in retrospective sounds a bit wonky.
The Power-Adapter is a 200W Power Brick.

How can I power the HDDs in a better way? It seems that I need to build some breakout cable myself?

Is 200W even enough for 5 HDDs, 1 SSD, 1NVMe?

31 minutes ago, SlidingGeneral said:

Is 200W even enough for 5 HDDs, 1 SSD, 1NVMe?

Not really. Consider that during a Parity Check, all 5 HDD are spun up and continuously accessing as well as the CPU performing the parity calculations. Add what is likely a cheaply made power brick (vs an actual PC power supply) and power routed through traces on the system's PC board. Not a recipe for success.