Hi,
I'm not sure but this seems to be a lot worse in 2.5.x for some
reason; my logs are full of I/O errors, not ready's and other errors from
my CDROM drive that is playing audio CDs; I suspect at least some of it
is due to kscd trying to figure out if there is a CD in an empty drive.
Is there any way to selectively silence errors from one IDE device? It
would make it easier to spot real/important errors from hard drives and the
like.
Dave
---------------- Have a happy GNU millennium! ----------------------
/ Dr. David Alan Gilbert | Running GNU/Linux on Alpha,68K| Happy \
\ gro.gilbert @ treblig.org | MIPS,x86,ARM,SPARC,PPC & HPPA | In Hex /
\ _________________________|_____ http://www.treblig.org |_______/
On Sad, 2003-05-10 at 21:17, Dr. David Alan Gilbert wrote:
> Hi,
> I'm not sure but this seems to be a lot worse in 2.5.x for some
> reason; my logs are full of I/O errors, not ready's and other errors from
> my CDROM drive that is playing audio CDs; I suspect at least some of it
> is due to kscd trying to figure out if there is a CD in an empty drive.
That shouldnt be generating messages. Its more important to know why or
to see wtf its doing that generates them
* Alan Cox ([email protected]) wrote:
> On Sad, 2003-05-10 at 21:17, Dr. David Alan Gilbert wrote:
^^^
Are you running with a Welsh locale there?
> > Hi,
> > I'm not sure but this seems to be a lot worse in 2.5.x for some
> > reason; my logs are full of I/O errors, not ready's and other errors from
> > my CDROM drive that is playing audio CDs; I suspect at least some of it
> > is due to kscd trying to figure out if there is a CD in an empty drive.
>
> That shouldnt be generating messages. Its more important to know why or
> to see wtf its doing that generates them
OK, the full messages on starting kscd up are:
May 10 23:37:32 gallifrey kernel: end_request: I/O error, dev hde,
sector 0
May 10 23:37:32 gallifrey last message repeated 3 times
May 10 23:37:32 gallifrey kernel: hde: packet command error: status=0x51
{ DriveReady SeekComplete Error }
May 10 23:37:32 gallifrey kernel: hde: packet command error: error=0x50
May 10 23:37:32 gallifrey kernel: end_request: I/O error, dev hde, sector 0
May 10 23:37:32 gallifrey kernel: ATAPI device hde:
May 10 23:37:32 gallifrey kernel: Error: Illegal request -- (Sense
key=0x05)
May 10 23:37:32 gallifrey kernel: Parameter list length error --
(asc=0x1a, ascq=0x00)
May 10 23:37:32 gallifrey kernel: The failed "Mode Select 10" packet
command was:
May 10 23:37:32 gallifrey kernel: "55 10 00 00 00 00 00 00 00 00 00 00
00 00 00 00 "
May 10 23:37:33 gallifrey kernel: end_request: I/O error, dev hde, sector 0
May 10 23:38:04 gallifrey last message repeated 31 times
That is with no disc in the drive; the drive identifies itself as:
'Pioneer DVD-ROM ATAPIModel DVD-116 0109' and driver is
'ide-cdrom version 4.59-ac1' (in 2.5.69)
But to be honest, my observation is more general - I'm sure I've seen
similar messages from VMware on systems (this system does not have it)
and other things rattling the CD drive on older kernels. In general I
think I'd just like to tell IDE to be quiet about certain drives so it
makes it easier to spot serious errors in the logs.
Dave
---------------- Have a happy GNU millennium! ----------------------
/ Dr. David Alan Gilbert | Running GNU/Linux on Alpha,68K| Happy \
\ gro.gilbert @ treblig.org | MIPS,x86,ARM,SPARC,PPC & HPPA | In Hex /
\ _________________________|_____ http://www.treblig.org |_______/
On Sad, 2003-05-10 at 23:47, Dr. David Alan Gilbert wrote:
> * Alan Cox ([email protected]) wrote:
> > On Sad, 2003-05-10 at 21:17, Dr. David Alan Gilbert wrote:
> ^^^
> Are you running with a Welsh locale there?
Ydw.
> May 10 23:37:32 gallifrey kernel: ATAPI device hde:
> May 10 23:37:32 gallifrey kernel: Error: Illegal request -- (Sense
> key=0x05)
> May 10 23:37:32 gallifrey kernel: Parameter list length error --
> (asc=0x1a, ascq=0x00)
> May 10 23:37:32 gallifrey kernel: The failed "Mode Select 10" packet
> command was:
> May 10 23:37:32 gallifrey kernel: "55 10 00 00 00 00 00 00 00 00 00 00
> 00 00 00 00 "
> May 10 23:37:33 gallifrey kernel: end_request: I/O error, dev hde, sector 0
> May 10 23:38:04 gallifrey last message repeated 31 times
That is interesting. Someone sent a SCSI command the it really didn't
like. This isn't a dell 8100 or similar laptop is it btw ?
> That is with no disc in the drive; the drive identifies itself as:
> 'Pioneer DVD-ROM ATAPIModel DVD-116 0109'
> and other things rattling the CD drive on older kernels. In general I
> think I'd just like to tell IDE to be quiet about certain drives so it
> makes it easier to spot serious errors in the logs.
I guess people with raw drive access should learn to program as well. You
could play with drive->quiet I guess (I think its drive->quiet) but right
now the IDE layer has no notion of how severe an error is although it has
some idea who caused it. For 2.5.x passing quiet/loud in the taskfile is
a viable extension for 2.4 its not so clear how you would do it nicely.
* Alan Cox ([email protected]) wrote:
> That is interesting. Someone sent a SCSI command the it really didn't
> like. This isn't a dell 8100 or similar laptop is it btw ?
Nope, thats just a standard DVD; its plugged into one channel of a
Promise Ultra100 TX2; an Iomega Zip 100 is a slave on the same channel.
Everything works fine - the only problem is the log messages from audio
playing. /proc/ide/hde/settings is:
name value min max
mode
---- ----- --- --- ----
current_speed 0 0 70 rw
dsc_overlap 1 0 1 rw
ide-scsi 0 0 1 rw
init_speed 0 0 70 rw
io_32bit 0 0 3 rw
keepsettings 0 0 1 rw
nice1 1 0 1 rw
number 0 0 3 rw
pio_mode write-only 0 255 w
slow 0 0 1 rw
unmaskirq 0 0 1 rw
using_dma 0 0 1 rw
My full set of IDE stuff is:
On board AMD 766 ViperPlus IDE:
hda - IBM Deathstar 60GB
hdc - Memorex CD-RW with ide-scsi
Promise Ultra100 TX2
hde - the offending DVD ROM
hdf - Iomega ZIP 100 ATAPI
hdg - IBM Deathstar 60GB
> I guess people with raw drive access should learn to program as well. You
> could play with drive->quiet I guess (I think its drive->quiet) but right
> now the IDE layer has no notion of how severe an error is although it has
> some idea who caused it. For 2.5.x passing quiet/loud in the taskfile is
> a viable extension for 2.4 its not so clear how you would do it nicely.
I guess I was really after a /proc/ide/hd?/verbosity to shut the drive
up independent of what application was trying to talk to it.
Dave
---------------- Have a happy GNU millennium! ----------------------
/ Dr. David Alan Gilbert | Running GNU/Linux on Alpha,68K| Happy \
\ gro.gilbert @ treblig.org | MIPS,x86,ARM,SPARC,PPC & HPPA | In Hex /
\ _________________________|_____ http://www.treblig.org |_______/
On Sat, 2003-05-10 at 22:41, Alan Cox wrote:
> On Sad, 2003-05-10 at 21:17, Dr. David Alan Gilbert wrote:
> > Hi,
> > I'm not sure but this seems to be a lot worse in 2.5.x for some
> > reason; my logs are full of I/O errors, not ready's and other errors from
> > my CDROM drive that is playing audio CDs; I suspect at least some of it
> > is due to kscd trying to figure out if there is a CD in an empty drive.
>
> That shouldnt be generating messages. Its more important to know why or
> to see wtf its doing that generates them
>
I have a Toshiba DVD drive, that while quiet in 2.4, generates 3-4
messages just during kernel start and initial module loading ..
I can forward it if need be.
Regards,
--
Martin Schlemmer
On Sat, May 10 2003, Dr. David Alan Gilbert wrote:
> Hi,
> I'm not sure but this seems to be a lot worse in 2.5.x for some
> reason; my logs are full of I/O errors, not ready's and other errors from
> my CDROM drive that is playing audio CDs; I suspect at least some of it
> is due to kscd trying to figure out if there is a CD in an empty drive.
This should fix it.
===== drivers/ide/ide-cd.c 1.44 vs edited =====
--- 1.44/drivers/ide/ide-cd.c Wed May 7 19:34:30 2003
+++ edited/drivers/ide/ide-cd.c Wed May 14 10:16:09 2003
@@ -2070,6 +2070,7 @@
req.sense = sense;
req.cmd[0] = GPCMD_TEST_UNIT_READY;
+ req.flags |= REQ_QUIET;
#if ! STANDARD_ATAPI
/* the Sanyo 3 CD changer uses byte 7 of TEST_UNIT_READY to
--
Jens Axboe