A failed drive causes system to hang

[Jeremy Chadwick]

{snipping lots for brevity}

On Sun, Apr 14, 2013 at 12:28:30PM -0700, Jeremy Chadwick wrote:
> On Sun, Apr 14, 2013 at 12:34:46PM +0200, Radio m?odych bandytw wrote:
> > Sorry. I thought just the error was important. So here you are:
> > dmesg.boot:
> > http://pastebin.com/LFXPusMX
> 
> Thank you.  Please read everything I have written below before doing
> anything.
> 
> Based on this output, we can see the following:
> 
> * AHCI is actively in use, and is a slowly-becoming-infamous ATI IXP700
>   controller:
> 
>   ahci0: <ATI IXP700 AHCI SATA controller> port 0xb000-0xb007,0xa000-0xa003,0x9000-0x9007,0x8000-0x8003,0x7000-0x700f mem 0xf9fffc00-0xf9ffffff irq 19 at device 17.0 on pci0
> 
> * The system has 3 disks attached to this controller:
> 
>   ada0 at ahcich0 bus 0 scbus2 target 0 lun 0
>   ada0: <WDC WD15EARS-22MVWB0 51.0AB51> ATA-8 SATA 2.x device
>   ada0: 300.000MB/s transfers (SATA 2.x, UDMA6, PIO 8192bytes)
>   ada0: Command Queueing enabled
>   ada0: 1430799MB (2930277168 512 byte sectors: 16H 63S/T 16383C)
>   ada1 at ata0 bus 0 scbus6 target 0 lun 0
>   ada1: <WDC WD15EARS-22MVWB0 51.0AB51> ATA-8 SATA 2.x device
>   ada1: 150.000MB/s transfers (SATA, UDMA6, PIO 8192bytes)
>   ada1: 1430799MB (2930277168 512 byte sectors: 16H 63S/T 16383C)
>   ada2 at ata0 bus 0 scbus6 target 1 lun 0
>   ada2: <ST3640323AS SD13> ATA-8 SATA 2.x device
>   ada2: 150.000MB/s transfers (SATA, UDMA6, PIO 8192bytes)
>   ada2: 610480MB (1250263728 512 byte sectors: 16H 63S/T 16383C)
> 
> Let's talk about ada0 and ada1 first.

Hold up a minute -- I just noticed some key information here (see what
happens with big conflated threads?), and it sheds some light on my
concerns with AHCI vs. classic ata(4):

ada0 -- attached to ahcich0
ada1 -- attached to ata0 (presumably a "master" drive)
ada2 -- attached to ata0 (presumably a "slave" drive)

This is extremely confusing, because ata0 is a classic ATA controller (I
can even tell from the classic ISA I/O port ranges):

atapci1: <ATI IXP700/800 UDMA133 controller> port 0x1f0-0x1f7,0x3f6,0x170-0x177,0x376,0xff00-0xff0f at device 20.1 on pci0
ata0: <ATA channel> at channel 0 on atapci1
ata1: <ATA channel> at channel 1 on atapci1

Yet the WD15EARS and ST3640323AS drives are physically SATA drives.

Are you using SATA-to-IDE adapters on these two drives?

If not, this seems to indicate the motherboard and/or SATA controller
is actually only binding 1 disk to AHCI, while the others are bound to
the same controller operating in (possibly) "SATA Enhanced" mode.

This would be the first I've ever seen of this (a controller operating
in both modes simultaneously), but I have a lot more experience with
Intel SATA controllers than I do AMD.

I don't know why a system would do this, unless all of this can be
controlled via the BIOS somehow.  What a mess.

-- 
| Jeremy Chadwick                                   jdc at koitsu.org |
| UNIX Systems Administrator                http://jdc.koitsu.org/ |
| Mountain View, CA, US                                            |
| Making life hard for others since 1977.             PGP 4BD6C0CB |