Hi folks,
Any suggestions welcomed. I've got a Dell Latitude D400 which I have installed
NV24 on, apparently successfully; but when I try to run "format" I get nasty
kernel errors as soon as I select the hard disk:
ata_disk_iosetup: byte count 0
There are then a series of error messages at longish intervals (see attached
file for my comp.unix.solaris posting/plea for help, with more info and a log
of a serial console session with all the grisly details). Sometimes I get to
interact with format after a few minutes, at which time I can use "verify" to
see that it has indeed read the partition table. Other times, the machine
kernel panics - read the attachment for more on that front. I tried booting
into single user from the NV19 installation CD, and the same thing happens.
I would suspect media errors, but running "dd" reading across the entire drive
under both Linux and Solaris fails to show up any problem in that regard.
Anyway, if anyone has any suggestions, I'd be most grateful. I've tried the
obvious tweaks in ata.conf (disable DMA, setting the mysterious "some laptops
need this" timing_flags to 1) and they haven't helped. "truss" shows that the
problem is triggered during an IOCTL on one of the raw devices (can't remember
if it's c1d0p0 or c1d0s2 though), so maybe this is my big motivation for
learning DTrace to pull apart the argument structures. :-)
Cheers,
Jason
This message posted from opensolaris.org
Subject:
Solaris Express NV24 on a Dell D400: "format" causes disk errors, kernel panic.
From:
Jason Ozolins <[EMAIL PROTECTED]>
Date:
Tue, 08 Nov 2005 18:01:22 +1100
Newsgroups:
comp.unix.solaris
Hi folks,
I installed Solaris Express Community Edition, nv24 build, onto a Dell Latitude
D400 laptop, seemingly successfully. I can login, I can dd the entire raw disk
(c1d0p0) to /dev/null without seeing any media errors, but as soon as I run
format and select the hard disk, things go weird, with "ata_disk_iosetup: byte
count zero" errors, followed by assorted other disk-related errors. After a
long time, the ioctl system call that provoked the errors finally completes,
and format at least allows me to display the partitioning of the disk, if not
actually do anything.
I found the "timing_flags" entry in /platform/i86pc/kernel/drv/ata.conf, and
tried setting that to 1, based on the associated comment that "some laptops
need this"; it seemed to make the machine panic upon running format, but then I
changed the timing flag back, and the machine still panics running format, so
I'm not sure what happened there. Also tried disabling DMA in ata.conf, and
the error messages are much the same, minus the initial ata_disk_iosetup error.
Neither Solaris nor Knoppix pick up any media errors on the disk when I use dd
to read across the entire disk, so I don't think it's the drive itself.
A capture of the console output for the DMA enabled, timing_flags=1 case,
including the kernel panic stack trace, is included below; if any Sun folks
happen to read this and can offer advice, I'd be very grateful.
Cheers,
Jason =:^/
console log:
Last login: Tue Nov 8 18:06:40 on console
Nov 8 18:10:56 piglet login: ROOT LOGIN /dev/console
Sun Microsystems Inc. SunOS 5.11 snv_24 October 2007
-bash-3.00# format
Searching for disks...done
AVAILABLE DISK SELECTIONS:
0. c1d0 <DEFAULT cyl 20318 alt 2 hd 32 sec 63>
/[EMAIL PROTECTED],0/[EMAIL PROTECTED],1/[EMAIL PROTECTED]/[EMAIL
PROTECTED],0
Specify disk (enter its number): 0
selecting c1d0
Nov 8 18:11:02 piglet genunix: ata_disk_iosetup: byte count zero
Nov 8 18:11:37 piglet scsi: WARNING: /[EMAIL PROTECTED],0/[EMAIL
PROTECTED],1/[EMAIL PROTECTED] (ata0):
Nov 8 18:11:37 piglet timeout: abort request, target=0 lun=0
Nov 8 18:11:37 piglet scsi: WARNING: /[EMAIL PROTECTED],0/[EMAIL
PROTECTED],1/[EMAIL PROTECTED] (ata0):
Nov 8 18:11:37 piglet timeout: abort device, target=0 lun=0
Nov 8 18:11:37 piglet scsi: WARNING: /[EMAIL PROTECTED],0/[EMAIL
PROTECTED],1/[EMAIL PROTECTED] (ata0):
Nov 8 18:11:37 piglet timeout: reset target, target=0 lun=0
Nov 8 18:11:37 piglet scsi: WARNING: /[EMAIL PROTECTED],0/[EMAIL
PROTECTED],1/[EMAIL PROTECTED] (ata0):
Nov 8 18:11:37 piglet timeout: reset bus, target=0 lun=0
Nov 8 18:11:37 piglet scsi: WARNING: /[EMAIL PROTECTED],0/[EMAIL
PROTECTED],1/[EMAIL PROTECTED] (ata0):
Nov 8 18:11:37 piglet timeout: early timeout, target=0 lun=0
Nov 8 18:11:37 piglet gda: WARNING: /[EMAIL PROTECTED],0/[EMAIL
PROTECTED],1/[EMAIL PROTECTED]/[EMAIL PROTECTED],0 (Disk0):
Nov 8 18:11:37 piglet Error for command 'Unknown' Error Level:
Informational
Nov 8 18:11:37 piglet gda: Requested Block 47670337, Error Block: 47641609
Nov 8 18:11:37 piglet gda: Sense Key: aborted command
Nov 8 18:11:37 piglet gda: Vendor 'Gen-ATA ' error code: 0x3
Nov 8 18:11:37 piglet gda: WARNING: /[EMAIL PROTECTED],0/[EMAIL
PROTECTED],1/[EMAIL PROTECTED]/[EMAIL PROTECTED],0 (Disk0):
Nov 8 18:11:37 piglet Error for command 'write sector' Error Level:
Informational
Nov 8 18:11:37 piglet gda: Sense Key: aborted command
Nov 8 18:11:37 piglet gda: Vendor 'Gen-ATA ' error code: 0x3
Nov 8 18:11:37 piglet genunix: ata_disk_iosetup: byte count zero
WARNING: /[EMAIL PROTECTED],0/[EMAIL PROTECTED],1/[EMAIL PROTECTED] (ata0):
timeout: abort request, target=0 lun=0
WARNING: /[EMAIL PROTECTED],0/[EMAIL PROTECTED],1/[EMAIL PROTECTED] (ata0):
timeout: abort device, target=0 lun=0
WARNING: /[EMAIL PROTECTED],0/[EMAIL PROTECTED],1/[EMAIL PROTECTED] (ata0):
timeout: reset target, target=0 lun=0
WARNING: /[EMAIL PROTECTED],0/[EMAIL PROTECTED],1/[EMAIL PROTECTED] (ata0):
timeout: reset bus, target=0 lun=0
WARNING: /[EMAIL PROTECTED],0/[EMAIL PROTECTED],1/[EMAIL PROTECTED] (ata0):
timeout: early timeout, target=0 lun=0
WARNING: /[EMAIL PROTECTED],0/[EMAIL PROTECTED],1/[EMAIL PROTECTED]/[EMAIL
PROTECTED],0 (Disk0):
Error for command 'Unknown' Error Level: Informational
Requested Block 47670337, Error Block: 47641609
Sense Key: aborted command
Vendor 'Gen-ATA ' error code: 0x3
WARNING: /[EMAIL PROTECTED],0/[EMAIL PROTECTED],1/[EMAIL PROTECTED]/[EMAIL
PROTECTED],0 (Disk0):
Error for command 'write sector' Error Level: Informational
Sense Key: aborted command
Vendor 'Gen-ATA ' error code: 0x3
panic[cpu0]/thread=d461a000: BAD TRAP: type=e (#pf Page fault) rp=d50c19e0
addr=48 occurred in module "dadk" due to a NULL pointer d
ereference
format: #pf Page fault
Bad kernel fault at addr=0x48
pid=528, pc=0xfea32f5c, sp=0x0, eflags=0x10282
cr0: 8005003b<pg,wp,ne,et,ts,mp,pe> cr4: 6d8<xmme,fxsr,pge,mce,pse,de>
cr2: 48 cr3: 13c28000
gs: 9eb01b0 fs: d50c0000 es: 80160 ds: fec80160
edi: 0 esi: 0 ebp: d50c1a48 esp: d50c1a18
ebx: 0 edx: d461a000 ecx: d3fb9104 eax: d51d27a8
trp: e err: 0 eip: fea32f5c cs: 158
efl: 10282 usp: 0 ss: d51d27a8
d50c192c unix:die+98 (e, d50c19e0, 48, 0)
d50c19cc unix:trap+1169 (d50c19e0, 48, 0)
d50c19e0 unix:_cmntrap+9b (9eb01b0, d50c0000, )
d50c1a48 dadk:dadk_pktprep+17 (0, 0, d51d27a8, fea)
d50c1a7c dadk:dadk_dk+20 (0, d4c34f00, d51d27)
d50c1a98 dadk:dadk_dk_strategy+15 (d51d27a8, d51d27a8,)
d50c1b00 genunix:default_physio+4ea (fea336a4, d51d27a8,)
d50c1b24 genunix:physio+1d (fea336a4, d51d27a8,)
d50c1b88 dadk:dadk_dk_buf_setup+88 (dad1cb58, d4c34f00,)
d50c1bd0 dadk:dadk_ioctl+4d (dad1cb58, 1980002, )
d50c1e60 cmdk:cmdkioctl+be (1980002, 5, 8047ac4)
d50c1e8c genunix:cdev_ioctl+2e (1980002, 5, 8047ac4)
d50c1eb0 specfs:spec_ioctl+65 (d56029c0, 5, 8047ac)
d50c1ed4 genunix:fop_ioctl+21 (d56029c0, 5, 8047ac)
d50c1f84 genunix:ioctl+151 (3, 5, 8047ac4, d50c)
_______________________________________________
opensolaris-help mailing list
[email protected]
