adaptec 2120S report

[Scott Long]

On Fri, 19 Sep 2003, Andreas Braukmann wrote:
>
> -- Scott Long <scottl at freebsd.org> wrote:
>
> > I've having trouble reproducing this.  The most obvious question that
> > comes to mind is, which method of building the array are you using?
> > Creation via the 'Clear' method would likely cause the problems that
> > you are seeing.
>
> The 2200S in the box besides me is in the process of building
> a 20 GByte RAID 5 volume ("build/verify"-method).
> Booting -current (a rather old [ca. 2003/03] snapshot install
>  cd) fails. The kernel freezes  after announcing the presence
> of the raid-volume and two error messages:
>
> | [...]
> | aac0: <Adaptec SCSI RAID 2200S> mem [...] irq 11 at device 10 on pci0
> | aac0: i960RX 100MHz, 48MB cache memory, optional battery present
> | aac0: Kernel 4.0-0, Build 6003, S/N b8****
> | aac0: Suupported Options=1f7e<[...]>
> | [...]
> | aacd0: <RAID 5> on aac0
> | aacd0: 20480 MB (41943040 sectors)
> | aac0: Error 5 sending GetDeviceProbeInfo command
> | aac0: Error 5 sending GetDeviceProbeInfo command
> | [...]
>
> I hope, that I get a fresh -current on my test machine, to
> provide further testing information.
>
> [... a few minutes later ...]
>
> Booting from a 2003/09/13 current-snapshot-cd yields nearly the
> same result.
> | aac0 Error 5 sending GetDeviceProbeInfo command
> | [...]
> | COMMAND 0xc28d80a4 TIMEOUT AFTER 51 SECONDS
> | COMMAND 0xc28d80a4 TIMEOUT AFTER 116 SECONDS
> | [... and so on ...]
>

Disable the 'aacp' device from your kernel config.  It is known to be
dodgy.

Scott