Bug 143424

Summary: Cann´t record cds or dvds
Product: [openSUSE] SUSE LINUX 10.0 Reporter: Heiko Schmidt <heiko.schmidt>
Component: OtherAssignee: Vladimir Nadvornik <nadvornik>
Status: RESOLVED WONTFIX QA Contact: E-mail List <qa-bugs>
Severity: Normal    
Priority: P5 - None    
Version: unspecified   
Target Milestone: ---   
Hardware: x86-64   
OS: SuSE Linux 10.0   
Whiteboard:
Found By: Beta-Customer Services Priority:
Business Priority: Blocker: ---
Marketing QA Status: --- IT Deployment: ---
Attachments: lspci -v
hwinfo --disk
boot.msg
/var/log/messages
cat /proc/interrupts
hwinfo --cdrom
cdrecord dev=help
tail -50 /var/log/messages
perl -nle '(/APIC/) || print $_;' </var/log/messages
output from cdrecord

Description Heiko Schmidt 2006-01-16 23:35:13 UTC
Hello,

when I tried to record some *.wav - files on cd or *.vob - files on DVD my system stucks, It seems to cann´t use the cd-/dvd-writer in the wrigth way. Unmounting isn´t possible cause `device is busy`. Unmounting the device before using xcdroast or k3b doesn´t help too.
Comment 1 Heiko Schmidt 2006-01-16 23:36:09 UTC
Created attachment 63518 [details]
lspci -v
Comment 2 Michael Gross 2006-01-17 14:11:04 UTC
Probably UDMA is not enabled for these drives. Of course, while the write-process in in progress, an unmount is not possible. You can check this condidition with `hdparm -d /dev/hdc' -- supposably the drive in question is hdc. This should give you `using_dma    =  1 (on)' as output.
Comment 3 Michael Gross 2006-01-23 14:16:20 UTC
Please reopen this bug if the problem still exists and the information can be provided.
Comment 4 Michael Gross 2006-01-23 14:18:21 UTC
see comment #3
Comment 5 Heiko Schmidt 2006-01-24 12:10:11 UTC
hdparm -d /dev/hdc returns 'using_dma    =  1 (on)', but the problem is still alive.
Comment 6 Michael Gross 2006-01-24 18:05:05 UTC
How did you attach your drives? (hwinfo --disk)
Comment 7 Heiko Schmidt 2006-01-30 22:44:54 UTC
Hello,

please see attachment
Comment 8 Michael Gross 2006-01-31 10:20:23 UTC
I would if there was one ;)
Comment 9 Heiko Schmidt 2006-01-31 10:55:03 UTC
Created attachment 65828 [details]
hwinfo --disk
Comment 10 Heiko Schmidt 2006-01-31 10:55:49 UTC
Sorry, unfortunatly I put it in bug #132128. But here is it.
Comment 11 Michael Gross 2006-02-01 16:28:21 UTC
Are there any messages in the syslog (/var/log/messages) when writing CDs? Does this problem occurr with any writing program? (Please note that xcdrost and AFAIK also k3b are frontends for xcdroast). Please also attach your boot messages (/var/log/boot.msg).
Comment 12 Michael Gross 2006-02-01 16:36:57 UTC
It is possible your drive is connected to the same ide-channel, for example with your harddisk you're reading the image from? This would probably result in a stucking system, because the IDE system will make a lot of overhead. If so, use seperate channels.
Comment 13 Heiko Schmidt 2006-02-03 22:03:42 UTC
My disk is on /dev/hda and my dvd-drive on /dev/hdc so i think it should be on different channels
Comment 14 Heiko Schmidt 2006-02-03 22:05:23 UTC
Created attachment 66433 [details]
boot.msg
Comment 15 Heiko Schmidt 2006-02-03 23:01:58 UTC
Created attachment 66438 [details]
/var/log/messages
Comment 16 Michael Gross 2006-02-06 15:45:03 UTC
Please don't compress logfiles, it makes the handling more complicated.
There seems to be something awfully wrong with your system, as your boot.msg is flooded with APIC errors. Try booting with pci=routeirq and/or acpi=off. Does this help?
Comment 17 Heiko Schmidt 2006-02-07 20:43:45 UTC
Alas this doesn´t help. My laptop is only with the parameter noacpi and pci=noacpi bootable.
I only compressed the logfile because of your upload filesize limitation.
Comment 18 Michael Gross 2006-02-08 12:01:56 UTC
Please attach /proc/interrupts here.
Comment 19 Martin Lasarsch 2006-02-09 11:48:46 UTC
>when I tried to record some *.wav - files on cd or *.vob - files on DVD my
>system stucks,

what does that mean? Is the system freezing/crashing?

Burning other files work?
Comment 20 Heiko Schmidt 2006-02-09 13:04:29 UTC
No I cann´t burn anything

cat /proc/ioports will be delivered this evening
Comment 21 Michael Gross 2006-02-09 17:53:35 UTC
We asked for /proc/interrupts, not /proc/ioports
Comment 22 Heiko Schmidt 2006-02-09 21:47:35 UTC
Created attachment 67434 [details]
cat /proc/interrupts
Comment 23 Michael Gross 2006-02-13 10:40:43 UTC
Please give a more detailed answear to comment #19, the interrup configuration seems OK.
Comment 24 Heiko Schmidt 2006-02-13 11:17:26 UTC
The system stucks means I start burning, the applocation tells me it´s starting the burn process and then nothing happens. 
The only way to get out of this is to kill the application. But then the CD/DVD drive is still in use, so that I cann´t halt or reboot the system, which is necessary to use the drive in an other way.
Comment 25 Michael Gross 2006-02-13 16:21:45 UTC
Does it work if you use cdrecord or dvdrecord directly on the shell? Might as well be an issue with the interface.
Comment 26 Heiko Schmidt 2006-02-14 22:09:14 UTC
Trying cdrecord -scanbus results in:
cann´t open /dev/pg*
cann´t open SCSI drive
no possible target for -scanbus

more see attachments
Comment 27 Heiko Schmidt 2006-02-14 22:10:38 UTC
Created attachment 68457 [details]
hwinfo --cdrom
Comment 28 Heiko Schmidt 2006-02-14 22:11:22 UTC
Created attachment 68458 [details]
cdrecord dev=help
Comment 29 Michael Gross 2006-02-15 11:18:33 UTC
If your devices are not attached on a SCSI bus, the scan will of course not work. These tools support direct adressing since Kernel 2.6 like dev=/dev/hdc, try it that way if hdc is your writer:

$cdrecord dev=/dev/hdc speed=32 driveropts=burnfree -dao -v cd.iso

And post the output here.
Comment 30 Heiko Schmidt 2006-02-15 20:45:57 UTC
$ cdrecord dev=/dev/hdc speed=32 driveropts=burnfree -dao -v cd_image.iso
Cdrecord-Clone 2.01 (x86_64-suse-linux) Copyright (C) 1995-2004 Jörg Schilling
Note: This version is an unofficial (modified) version
Note: and therefore may have bugs that are not present in the original.
Note: Please send bug reports or support requests to http://www.suse.de/feedback
Note: The author of cdrecord should not be bothered with problems in this version.
TOC Type: 1 = CD-ROM
scsidev: '/dev/hdc'
devname: '/dev/hdc'
scsibus: -2 target: -2 lun: -2
Warning: Open by 'devname' is unintentional and not supported.
Linux sg driver version: 3.5.27
Using libscg version 'schily-0.8'.
cdrecord: Warning: using inofficial libscg transport code version (okir@suse.de-scsi-linux-sg.c-1.83-resmgr-patch '@(#)scsi-linux-sg.c  1.83 04/05/20 Copyright 1997 J. Schilling').
Driveropts: 'burnfree'
SCSI buffer size: 64512
atapi: 1
Device type    : Removable CD-ROM
Version        : 0
Response Format: 2
Capabilities   :
Vendor_info    : 'HL-DT-ST'
Identifikation : 'DVD-RW GWA-4082N'
Revision       : 'CL02'
Device seems to be: Generic mmc2 DVD-R/DVD-RW.
Current: 0x0009
Profile: 0x0011
Profile: 0x0015
Profile: 0x0014
Profile: 0x0013
Profile: 0x001A
Profile: 0x001B
Profile: 0x002B
Profile: 0x0010
Profile: 0x0009 (current)
Profile: 0x000A
Profile: 0x0008


after this nothing else happens
Comment 31 Michael Gross 2006-02-16 12:18:18 UTC
I see, so the problem has most likely nothing to do with xcdroast or k3b.
Can you see any access at your drive (does the LED flicker)?

Please run the same command with debug=2 and kdebug=2 then attach the output as file and 500 lines of your /var/log/messages.
Comment 32 Michael Gross 2006-02-22 10:05:13 UTC
Please reopen this report once you can provide the required information.
Comment 33 Heiko Schmidt 2006-02-22 22:17:22 UTC
dev: '/dev/hdc' speed: 32 fs: -1 driveropts 'burnfree'
Cdrecord-Clone 2.01 (x86_64-suse-linux) Copyright (C) 1995-2004 Jörg Schilling
Note: This version is an unofficial (modified) version
Note: and therefore may have bugs that are not present in the original.
Note: Please send bug reports or support requests to http://www.suse.de/feedbackNote: The author of cdrecord should not be bothered with problems in this version.
TOC Type: 0 = CD-DA
cdrecord: Operation not permitted. WARNING: Cannot set RR-scheduler
cdrecord: Permission denied. WARNING: Cannot set priority using setpriority().
cdrecord: WARNING: This causes a high risk for buffer underruns.
fs: 4194304 buflen: 4198400
cdrecord: shared memory segment attached at: 2AAAAAEF3000 size 4198400
buf: 2AAAAAEF3000 bufend: 2AAAAB2F4000, buflen: 4198400
buf: 2AAAAAEF3000 bufend: 2AAAAB2F4000, buflen: 4198400 (align 0)
cdrecord: Warning: not running as root user, fs= option ignored.
scsidev: '/dev/hdc'
devname: '/dev/hdc'
scsibus: -2 target: -2 lun: -2
scg__open(/dev/hdc) -2,-2,-2
Warning: Open by 'devname' is unintentional and not supported.
Linux sg driver version: 3.5.27
Using libscg version 'schily-0.8'.
cdrecord: Warning: using inofficial libscg transport code version (okir@suse.de-scsi-linux-sg.c-1.83-resmgr-patch '@(#)scsi-linux-sg.c  1.83 04/05/20 Copyright 1997 J. Schilling').
Driveropts: 'burnfree'
Target (1,0,0): DMA max 129024 old max: 64512
SCSI buffer size: 64512
Target (1,0,0): DMA max 129024 old max: 64512
scgo_getbuf: 64512 bytes
atapi: 1
ioctl ret: 0
host_status: 00 driver_status: 00
ioctl ret: 0
host_status: 00 driver_status: 00
ioctl ret: 0
host_status: 00 driver_status: 00
ioctl ret: 0
host_status: 00 driver_status: 00
ioctl ret: 0
host_status: 00 driver_status: 00
ioctl ret: 0
host_status: 00 driver_status: 00
ioctl ret: 0
host_status: 00 driver_status: 00
ioctl ret: 0
host_status: 00 driver_status: 00
ioctl ret: 0
host_status: 00 driver_status: 00
ioctl ret: 0
host_status: 00 driver_status: 00
ioctl ret: 0
host_status: 00 driver_status: 00
ioctl ret: 0
host_status: 00 driver_status: 00
Device type    : Removable CD-ROM
Version        : 0
Response Format: 2
Capabilities   :
Vendor_info    : 'HL-DT-ST'
Identifikation : 'DVD-RW GWA-4082N'
Revision       : 'CL02'
Device seems to be: Generic mmc2 DVD-R/DVD-RW.
ioctl ret: 0
host_status: 00 driver_status: 00
ioctl ret: 0
host_status: 00 driver_status: 00
ioctl ret: 0
host_status: 00 driver_status: 00
ioctl ret: 0
host_status: 00 driver_status: 00
ioctl ret: 0
host_status: 00 driver_status: 00
ioctl ret: 0
host_status: 00 driver_status: 00
ioctl ret: 0
host_status: 00 driver_status: 00
ioctl ret: 0
host_status: 00 driver_status: 00
ioctl ret: 0
host_status: 00 driver_status: 00
ioctl ret: 0
host_status: 00 driver_status: 00
ioctl ret: 0
host_status: 00 driver_status: 00
ioctl ret: 0
host_status: 00 driver_status: 00
ioctl ret: 0
host_status: 00 driver_status: 00
ioctl ret: 0
host_status: 00 driver_status: 00
ioctl ret: 0
host_status: 00 driver_status: 00
Current: 0x0009
Profile: 0x0011
Profile: 0x0015
Profile: 0x0014
Profile: 0x0013
Profile: 0x001A
Profile: 0x001B
Profile: 0x002B
Profile: 0x0010
Profile: 0x0009 (current)
Profile: 0x000A
Profile: 0x0008

Comment 34 Michael Gross 2006-02-23 10:36:30 UTC
Is there nothing in the syslog?
Does it change anything if you start this process as root?
Comment 35 Heiko Schmidt 2006-02-23 22:52:23 UTC
I started the process either as root or as normal user with the same result.
My /var/log/messages is spamed by APIC errors.
So I attach a tail -50 /var/log/messages and
perl -nle '(/APIC/) || print $_;' < /var/log/messages.
But I think there ´s no further information.
Comment 36 Heiko Schmidt 2006-02-23 22:54:12 UTC
Created attachment 70064 [details]
tail -50 /var/log/messages
Comment 37 Heiko Schmidt 2006-02-23 22:55:15 UTC
Created attachment 70065 [details]
perl -nle '(/APIC/) || print $_;' </var/log/messages
Comment 38 Michael Gross 2006-02-24 11:58:53 UTC
Does booting with pci=routirq or pci=noacpi or maby acpi=off solve the problem? Obviously this is an interrupt problem maby caused by a buggy BIOS.
Comment 39 Heiko Schmidt 2006-02-24 12:28:47 UTC
I just tried booting with pci=routirq to get rid of the spaming ACIP errors but didn´t work. My system only boots with pci=noacpi  acpi=off. I will add pci=routirq too and try burning again.
Is there any possibility of changing the irq after boottime ?
Comment 40 Heiko Schmidt 2006-02-25 19:42:37 UTC
Alas this doesn´t change anything, you have another idea ?
During the attempt of burning, I tried hwinfo --cdrom and return stucks.
Comment 41 Michael Gross 2006-02-27 09:54:43 UTC
Vladimir: We need your help. Any suggestion about how to debug this - if it is a bug at all?
Comment 42 Vladimir Nadvornik 2006-02-27 10:21:33 UTC
You can get even more verbose output from cdrecord by:

cdrecord dev=/dev/hdc speed=32 driveropts=burnfree -dao -v -V -d cd_image.iso



Comment 43 Heiko Schmidt 2006-03-02 00:11:08 UTC
The result is attached. 
Comment 44 Heiko Schmidt 2006-03-02 00:11:55 UTC
Created attachment 70890 [details]
output from cdrecord
Comment 45 Vladimir Nadvornik 2006-03-15 14:31:22 UTC
It freezes on the 'read dvd structure' command.
This command is used on DVD burners to detect whether the media is CD or DVD.
It is the first command that actually accesses the media.

It is possible that the burner does not like the media. Maybe you can try 
another media brand. Firmware upgrade could help too.

Did you try burning with any other OS?



Comment 46 Heiko Schmidt 2006-03-15 14:35:52 UTC
Burning with WinXP and Nero works fine with the same media.
I will look for an upgrade.
Comment 47 Vladimir Nadvornik 2006-03-15 15:10:51 UTC
Jens, could it be anything in kernel? Is there anything special on 'read dvd structure' command?
Comment 48 Jens Axboe 2006-03-16 20:31:51 UTC
I don't think it's the kernel, there's nothing special about that commands. Try disabling dma with hdparm -d0 /dev/hdc, it might be that the drive doesn't like the alignment. Did previous kernels work, or is this the first you tried?

Also, sometimes the problematic command isn't the last one but rather one issued before that confused the drive. So it might not be 'read dvd structure', but something before that.
Comment 49 Heiko Schmidt 2006-03-22 20:49:03 UTC
At the beginning it seems to work but then it freezes my system completely.
I tried it two times withthe same result. It starts burning for some percent of the first track and then it freezes.
Comment 50 Heiko Schmidt 2006-03-24 14:37:50 UTC
This is the first kernel I tried on this notebook.
Comment 51 Vladimir Nadvornik 2007-04-24 12:45:52 UTC
There were some fixes since 10.0. There is a high chance that it is already fixed
on newer distros.
Please reopen this bug if it happens on 10.2 or factory.