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

Re: Ominous looking syslog messages



On Thursday 17 February 2005 17:18, 
debian-user-digest-request@lists.debian.org wrote:
> > Feb 17 12:25:29 d_baron kernel: hdb: status error: status=0x00 { }
> > Feb 17 12:25:29 d_baron kernel: hdb: status error: error=0x00
> > Feb 17 12:25:34 d_baron kernel: hda: status timeout: status=0x80 { Busy }
> > Feb 17 12:25:34 d_baron kernel:
> > Feb 17 12:25:34 d_baron kernel: hda: DMA disabled
> > Feb 17 12:25:34 d_baron kernel: hda: drive not ready for command
> > Feb 17 12:25:34 d_baron kernel: ide0: reset: success
> >
> > There was nothing immediately before these indicating what initiated
> > whatever brought this on.
> >
> > /dev/hda's DMA was indeed set off. hparm -d1 put it on again.
> > hdb is a CD drive. No disk in it
> > hda has two vfat partitions on it for Windows
> >
> > Linux in on hdc (which also has one vfat). There is a second CD on hdd
> > Disks are of recent vintage.
>
> Which IDE controllers are your devices connected to?  You have /dev/hda
> and the CD drive on the primary controller and /dev/hdc HDD and /dev/hdd
> CD on the secondary controller?
>
> This sounds like a non-optimal setup - normally you get better
> performance by putting the HDDs on the first controller and the two CDs
> on the second controller.
This is how they are connected. This controversy goes on and on. I have not 
really noticed differences in performance from the "old way". I was advised 
to do it this way because I need throughput between the two HDs for audio 
production, and that the problem of mixing HDs and CDs on an IDE bus is long 
ago irrelevant. Who knows?

>
> However, that in itself doesn't explain your errors.  Sometimes you get
> errors like that if you try to mount a CD which is not mountable (audio
> CD) or if the CD is damaged (which might explain the hdb errors).
It looks as if something was trying to access that CD drive had no disk in it. 
I did not try to access it so the question in what?

> If the errors keep occuring and/or become more frequent, it could
> suggest a hardware issue: IDE controller or one of the drives.
!



Reply to: