Temperature sensor on SCSI disks (IBM / Hitachi)

[Scott Long]

Panagiotis Christias wrote:
> On 3/30/06, Oliver Fromme <olli at lurza.secnetix.de> wrote:
> 
>>Kenneth D. Merry <ken at freebsd.org> wrote:
>> > Oliver Fromme wrote:
>> > > I have the following SCSI disks in a server:
>> > >
>> > > <IBM DDYS-T18350M S96H>   at scbus0 target 0 lun 0 (pass0,da0)
>> > > <IBM DDYS-T18350M S96H>   at scbus0 target 1 lun 0 (pass1,da1)
>> > >
>> > > Searching the mailing lists revealed that IBM SCSI disks
>> > > (Hitachi nowadays) have a temperature sensor that can be
>> > > queried with a special (prioprietary) command like this:
>> > > [...]
>> > With some more recent IBM drives (possibly including DDYS drives, can't
>> > remember), you can get the temperature like this:
>> >
>> > camcontrol cmd da0 -v -u $i -c "4D 0 6F 0 0 0 0 0 20 0" -i 32 "s10 i1"`
>>
>>Cool, that one works.
>>
>># camcontrol cmd da0 -v -c "4D 0 6F 0 0 0 0 0 20 0" -i 32 "s10 i1"
>>33
>># camcontrol cmd da1 -v -c "4D 0 6F 0 0 0 0 0 20 0" -i 32 "s10 i1"
>>36
>>
>>Thanks very much!
>>
>>Do such commands exist for other drives or vendors, too?
>>
>>Best regards
>>   Oliver
> 
> 
> Very interesting! I am wondering if it would be possible to get that
> kind of information if the disks are part of a hardware RAID. Our
> systems have IBM/Hitachi disks connected on RAID controllers
> (accessible using the amr driver). Any ideas?
> 
> Thank you in advance,
> Panagiotis

The amr driver has a SCSI passthrough mechanism for talking to the
underlying disks (assuming that the disks are SCSI and not SATA), but
it's been broken for several months.  Since such a mechanism has such
limited use, I haven't bothered to fix it.  I definitely don't have time
to fix it now, but I can give pointers to anyone who does want to look
at it.

Scott