CAM error on SSD

TrueNAS General
1

I recently had to change the boot ssd on my core NAS… I used a brand new Samsung EVO 870 500GB (ridiculously too big). The hardware is a HP MIcroserver N40L thats been running rock solid for 10 years now.

I actually installed a different replacement SSD (Intel SSD DC S3510) via a clean install and config restore, then found this ‘new’ Intel SSD had 55800 hours on it and a huge write count, ugh. So I got this brand new Samsung, plugged it in via a USB SATA cable, added it as a mirror, then shut down and swapped it into the internal SATA port. I detached the no-longer-existant Intel, and did a scrub. scrub showed no errors but the console barfed out a bunch of…

May 14 22:00:21 hognas (ada4:ahcich5:0:0:0): READ_FPDMA_QUEUED. ACB: 60 90 48 18 e8 40 00 00 00 00 00 00
May 14 22:00:21 hognas (ada4:ahcich5:0:0:0): CAM status: ATA Status Error
May 14 22:00:21 hognas (ada4:ahcich5:0:0:0): ATA status: 41 (DRDY ERR), error: 84 (ICRC ABRT )
May 14 22:00:21 hognas (ada4:ahcich5:0:0:0): RES: 41 84 48 18 e8 00 00 00 00 90 00
May 14 22:00:21 hognas (ada4:ahcich5:0:0:0): Retrying command, 3 more tries remain

and each time I scrub boot-pool, I get a bunch (100 or so?) more of these, again no errors shown in zpool status.

The Microserver N40L has a
00:11.0 SATA controller: Advanced Micro Devices, Inc. [AMD/ATI] SB7x0/SB8x0/SB9x0 SATA Controller [AHCI mode] (rev 40) (prog-if 01 [AHCI 1.0])
Subsystem: Hewlett-Packard Company ProLiant MicroServer N36L
Flags: bus master, 66MHz, medium devsel, latency 64, IRQ 19
I/O ports at d000
I/O ports at c000
I/O ports at b000
I/O ports at a000
I/O ports at 9000
Memory at fe6ffc00 (32-bit, non-prefetchable)
Capabilities: [50] MSI: Enable+ Count=1/8 Maskable- 64bit+
Capabilities: [70] SATA HBA v1.0
Capabilities: [a4] PCI Advanced Features

and I did see some very vague forum ramblings about older AMD SATA ports and Samsung SSD not getting along well, but nothing conclusive. This same system has run for years with a Crucial BX500, and it ran for a few days with the Intel DC S3510, without any errors.

2

Eh, no idea. Swap it with another disk/port and see if it still gives the same errors.

That’ll at least narrow the issue down to the disk or the port.

1 Like
3

thats not so easy on the Microserver, the 4 HD 3.5" bays are on a PCBA, while I’m running the SSD on the single other SATA port via a standard sata cable.

4

My experience is that CAM/etc errors are either cable or SATA controller/drive firmware issues. I’d try the following: 1) swap the SATA cable, 3) try a different SSD (small SSD’s are under ~$20), or, if the server has a PCI slot, a small, known good SATA card (watch for speific chipset), also under ~$20). ALso, in mt expeience, CAM errors, especially from the boot pool, have not created bigger issues. Good luck. John