[286] in linux-scsi channel archive
scsi problems with debian
daemon@ATHENA.MIT.EDU (Stuart Marshall)
Thu Jun 22 20:35:44 1995
Date: Thu, 22 Jun 95 16:46:17 PDT
From: Stuart Marshall <stuart@lensing.physics.ucsb.edu>
To: linux-scsi@vger.rutgers.edu
Hi,
I am having trouble that seems to be related to the ncr driver. I am in
the process of trying to upgrade to the debian distribution which has
a 1.2.1 kernel from a slackware distribution from sept 94 which had a
1.1.19 kernel. Since I started trying to install the new distribution
I have had the system freeze up multiple times. As near as I can tell
it is a scsi problem. I have a gigabyte pci 486 motherboard and an ncr53c810
pci scsi card that came with the pc (it is about 14 months old). I used
slackware (1.?) since september 94 and a 1.1.19 kernel that came with it.
I never once had any error messages or freeze ups of any kind (thanks Drew)
so I thought I was lucky given all I've read about ncr scsi problems with other
motherboards. The bios on the board is an ami with very few (none?) options
regarding the scsi system configuration and very few options in general.
So now I'm trying to use the 1.2.1 kernel that the installation and base
disks have. The system has died several times while formating the disk,
while loading the base 1,2,3 disks, while running dpkg and while just
sitting idle. When there are dying messages they usually say something
about the swapper and the scsi system. The kernel on these disks does
have the EATA scsi stuff and someone mentioned that may be related to the
problem. (I'm not sure what EATA does.)
At the end of this message are parts of two recent death sequences. The
first occured overnight on an otherwise idle system and the second occured
while running dpkg --auto --configure (lots of disk access).
Does anyone know whether this is a kernel problem, and scsi driver problem,
or a hardware problem? Would it be better with the 1.2.10 kernel?
I have most of the 0.93R5 packages installed and the system
limps along for random amounts of time. I can provide more complete error
messages if it would help.
thanks for listening,
Stuart Marshall
###
###------------ death on idle system overnight ----------------------------
###
# Kernel panic: EXT2-fs panic (device 8/2): read_block_bitmap: Cannot read block bitmap - block_group = 31, block_bitmap = 253956
# Kernel panic: EXT2-fs panic (device 8/2): ext2_write_inode: unable to read i-node block - inode = 24771, block= 98346
.
.
.
###
###------------ death while running dpkg -----------------------------------
###
#(lots of stuff that went by too quick to read with stack values and addresses)
.
.
.
Aiee, Killing interupt handler
Kfree of non-kmalloced memory: 001d5598, next= 00000000, order = 0
task[0] (swapper) killed: unable to recover
Kernel panic: Trying to free up swapper memory space
In swapper task - not syncing
scsi : aborting command due to timeout: pid 13126,
scsi0, id 0, lun 0, Write (6) 05 73 29 02 00
scsi: DANGER: command in running list, can not abort
scsi : aborting command due to timeout: pid 13126,
scsi0, id 0, lun 0, Write (6) 05 73 29 02 00
scsi: DANGER: command in running list, can not abort
SCSI host 0 abort() timed out - resetting
scsi0: DANGER: NCR53c7xx_reset is NOP
Unable to reset scsi host 0 - probably a SCSI bus hang
###---------------------- that's all ------------------------------------
