[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: hda: DMA timeout error, is it a problem ?



On Thu, Aug 23, 2007 at 10:12:33AM -0700, Bob McGowan wrote:
> Douglas A. Tutty wrote:
> >On Wed, Aug 22, 2007 at 08:49:34PM +0200, Shams Fantar wrote:
> >>I ask me if this message is a problem or not :
> >>
> >>Aug 22 11:03:20 sithare kernel: hda: dma_timer_expiry: dma status == 0x61
> >>Aug 22 11:03:20 sithare kernel: hda: DMA timeout error
> >>Aug 22 11:03:20 sithare kernel: hda: dma timeout error: status=0x58 { 
> >>DriveReady SeekComplete DataRequest Aug 22 11:03:20 sithare kernel: ide: 
> >>failed opcode was: unknown
 >
> >Try shutdown, disable DMA in the bios, and reboot into single-user mode
> >(avoids mounting the drive rw).  See if the messages show up in dmesg.
> >Then reboot into normal mode and check syslog for that boot (ignore
> >previous boots).  
> 
> I don't believe the read-only state is necessarily true.  As I 
> understand it (which could be wrong...), the root partition is mounted 
> read-only so fsck can run and fix minor problems (if any).  If that 
> fails, you're dumped into a single user shell, to run fsck manually, and 
> with root still mounted read-only.
> 
> But if the fsck is OK, root is remounted read/write, so you can do 
> system maintenance work from single user (which can include editing or 
> deleting files from the root partition).


Yes, I see that things get mounted in /etc/rcS.d before init even gets
to /etc/rc1.d

Try kernel command line with init=/bin/sh.

The initrd should ensure that you get a / mounted ro.

Or boot the install CD in rescue mode and go to a shell.

Doug.





Reply to: