[kwlug-disc] DRDY errors on drive that passes smartcontrol

[Andrew Kohlsmith (mailing lists account)]

> On Dec 28, 2015, at 7:46 PM, Paul Nijjar <paul_nijjar at yahoo.ca> wrote:
> Dec 28 15:21:18 my-server kernel: [3753525.630476] ata2.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x0
> Dec 28 15:21:18 my-server kernel: [3753525.695122] ata2.00: irq_stat 0x40000001
> Dec 28 15:21:19 my-server kernel: [3753525.757481] ata2.00: failed command: READ DMA
> Dec 28 15:21:19 my-server kernel: [3753525.819524] ata2.00: cmd c8/00:08:08:29:00/00:00:00:00:00/e0 tag 12 dma 4096 in
> Dec 28 15:21:19 my-server kernel: [3753525.819524]          res 61/04:08:08:29:00/00:00:00:00:00/e0 Emask 0x1 (device error)
> Dec 28 15:21:19 my-server kernel: [3753526.083712] ata2.00: status: { DRDY DF ERR }
> Dec 28 15:21:19 my-server kernel: [3753526.083713] ata2.00: error: { ABRT }

I agree. Don’t trust the drive and get any data that you haven’t already got off of it now. I don’t think the SATA Port is bad; it just seems to be reporting that a command that was issued returned a bad result. I’m not sure I’ve ever had a bad SATA controller or port, but perhaps someone with more servers under their belt could provide a smarter (ha ha ha) answer.

> but when I network boot the machine into a Live CD and run
> gsmartcontrol, both the short and extended tests pass. I can also fsck
> the file systems successfully. 

It has been only very rarely that I’ve had any warning from SMART that a drive was about to die. I’ve run drives that throw SMART "FAILING NOW” errors for years without a problem, and every drive that *has* died has had a clean bill of health according to SMART. Go figure.

The only thing I use it for is IDing the drives with the case on, and I run smartd more out of sheer blind determination to have the damn thing warn me correctly JUST ONCE before I die.

-A.