AFAIK, no one's responded to me about whether I should just turn off smartd.
At any rate, here's more info: this server only started spitting Apr 25 07:47:50 <server> kernel: sda: Current [descriptor]: sense key: Recovered Err ror Apr 25 07:47:50 <server> kernel: Add. Sense: ATA pass through information availl able Apr 25 07:47:50 <server> kernel: Apr 25 07:47:50 <server> kernel: Descriptor sense data with sense descriptors (in hh ex): Apr 25 07:47:50 <server> kernel: 72 01 00 1d 00 00 00 0e 09 0c 00 00 00 00 00 00 Apr 25 07:47:50 <server> kernel: 00 4f 00 c2 40 50
for every drive - and it has four - right after updated to the most current kernel 2.6.18-308.4.1.el5 #1 SMP and rebooted.
So one question becomes three: a) should I turn off smartd? b) should I reboot to the previous kernel, and see if it keeps up? c) is this a known bug?
mark