disk spin down <summary II>

[Peter S Galbraith]

Just a followup in case someone wants to follow my tracks...

Using Linux-2.0.31-pre-release-3 with the scsi-idle-1.3.98.patch,
I used the scsi-idle deamon to spin-down a disk which held an
enabled but rarely used swap partition.

After a weekend turned-on, my system was acting strangely this Monday
morning (window manager wouln't focus windows, fortune command would give
garbage, etc...).  So I rebooted and the system frooze on `turning swap
off' because the disk was spun-down.

After fixing the filesystem (breakage which could have by the unclean 
reboot, although I sync'ed first), the system was okay.

Since the problem could be due to other changes in 2.0.31-p3,
I rebuilt 2.0.30 with the patches (the 20-* file are from www.linuxhq.com):

 - aic7xxx-2.1.26-Jun1.tgz
 - 20-aic7-proc-overflow-fix.patch
 - 20-remove-disk.patch   (Thanks to JRPerkins at aol.com for pointing it out)
 - scsi-idle-1.3.98.patch

I am currently *not* running the swap partition on the spun-down disk.
I will report problems if any show up.  The following message appeared when
I called scsi-idle to spin-down the disk, but nothing since:

scsi : aborting command due to timeout : pid 4702, scsi0, channel 0, id 1, lun 0 Start/Stop Unit 00 00 00 00 00 
general protection: 0000
CPU:    0
EIP:    0010:[<00000004>]
EFLAGS: 00010202
eax: 00000000   ebx: 002fb214   ecx: 001d6ddc   edx: 03b20003
esi: 00000297   edi: 00000000   ebp: 00009410   esp: 001d4f14
ds: 0018   es: 0018   fs: 002b   gs: 0018   ss: 0018
Process swapper (pid: 0, process nr: 0, stackpage=001d3030)
Stack: 00196c43 002fb214 002fb214 00009410 00000206 001d4f68 00195794 002fb214 
       00000003 002fb214 001970fa 002fb214 00040000 001fd700 00000001 00112040 
       00000001 ffffffff 00000001 00000001 001d4f84 001fd570 001178ef 001d4f84 
Call Trace: [con_adjust_height+99/432] [qic02_tape_open+292/916] [vesa_unblank+98/252] [process_timeout+84/116] [sys_sigprocmask+35/164] [do_signal+119/636] [init+440/868] 
       [do_signal+241/636] [do_shell+64/196] [getphase+92/168] 
Code: d0 9b 00 f0 c3 e2 00 f0 d0 9b 00 f0 d0 9b 00 f0 54 ff 00 f0 
Aiee, killing interrupt handler
kfree of non-kmalloced memory: 001d5078, next= 00000246, order=0
kfree of non-kmalloced memory: 001d5068, next= 00000246, order=0
kfree of non-kmalloced memory: 001d557c, next= 00000246, order=0
idle task may not sleep
sage repeated 4 times
SCSI host 0 abort (pid 4702) timed out - resetting
SCSI bus is being reset for host 0 channel 0.
(scsi0:1:0) Abort_reset, scb flags 0x1, while idle, LASTPHASE = 0x1, SCSISIGI 0x0, SEQADDR 0x6, SSTAT0 0x5, SSTAT1 0xa
(scsi0:1:0) Queueing an Abort SCB.
(scsi0:1:0): Abort message sent.
SCSI host 0 channel 0 reset (pid 4702) timed out - trying harder
SCSI bus is being reset for host 0 channel 0.
(scsi0:-1:0) Reset device, active_scb 6
scsi0: (targ -1/chan A) matching scb to (targ 1/chan A)
scsi0: (targ -1/chan A) matching scb to (targ 1/chan A)
scsi0: Resetting current channel A
scsi0: Channel reset, sequencer restarted
(scsi0:1:0) Aborting scb 6

If the system runs cleanly after a few days, I may try with swap on again.
I'll let the list know how it goes.
--
Peter Galbraith, research scientist             <galbraith at mixing.qc.dfo.ca>
Maurice-Lamontagne Institute, Department of Fisheries and Oceans Canada
P.O. Box 1000, Mont-Joli Qc, G5H 3Z4 Canada  418-775-0852 - FAX 418-775-0546