First I had three messages in syslog (manual copy):
21:00:38 scsi: aborting command due to timeout: pid 48677,
scsi0, channel 0, id 0, lun 0 Read(6) 0f f0 2c 04 00
21:00:42 scsi: aborting command due to timeout: pid 48678,
scsi0, channel 0, id 0, lun 0 Wrute (6) 0c 08 80 08 00
21:00:44 scsi: aborting command due to timeout: pid 48679,
scsi0, channel 0, id 0, lun 0 Read(6) 0c 08 88 04 00
And on the console I had (kernel dubug flag set, klogd -c8 after fix)
SCSI : aborting command due to timeout: pid 48677, scsi 0, channel 0, id 0,
lun 0 Read (6) 0f f0 2c 04 00
SCSI host 0 abort (pid48677) timed out - resetting
SCSI bus is being reset for host 0 channel 0.
(scsi=:0:0) Abort_reset, scb flags 0x1, Message-In phase, SCSISIGI 0x6,
SEQADDR 0xcd, SSTAT0 0x7, STAT1 0x3
scsi0: Invalid SCB ID 7 is active, SCB falgs = 0x1.
(scsi0:-1:0) Reset device, active_scb 1
(scsi0: (targ -1/chan A) matching scb to (targ 0/Chan A)
(scsi0: (targ -1/chan A) matching scb to (targ 0/Chan A)
(scsi0: (targ -1/chan A) matching scb to (targ 0/Chan A)
Kernel panic: aic7xxx: Waiting list inconsistency; SCB index = 255, numscbs = 9
In swapper task - not syncing
The driver has tagged queueing enabled for the harddisk. Tape and disk
are doinig sync. transfers at 10 and 5MB/s.
The system was practically idle (I changed between playing nethack and
watching the eclipse of the moon (bright night; full-moon))
-- Ulrich Windl <Ulrich.Windl@rz.uni-regensburg.de>