CD write error:

bastl · 01-22-2012, 06:51 AM

Brand new dvd-writer LG-DVDRAM GH22NS50

But it does not burn:

Quote:

bash-3.2# cdrecord dev=6,0,0 speed=4 -raw96r -overburn driveropts=burnfree /home/bastl/img.iso
Cdrecord-ProDVD-ProBD-Clone 3.01a03 (i686-pc-linux-gnu) Copyright (C) 1995-2010 Joerg Schilling
scsidev: '6,0,0'
scsibus: 6 target: 0 lun: 0
Linux sg driver version: 3.5.34
Using libscg version 'schily-0.9'.
Device type : Removable CD-ROM
Version : 5
Response Format: 2
Capabilities :
Vendor_info : 'HL-DT-ST'
Identifikation : 'DVDRAM GH22NS50 '
Revision : 'TN03'
Device seems to be: Generic mmc2 DVD-R/DVD-RW/DVD-RAM.
Using generic SCSI-3/mmc CD-R/CD-RW driver (mmc_cdr).
Driver flags : MMC-3 SWABAUDIO BURNFREE
Supported modes: TAO PACKET SAO SAO/R96P SAO/R96R RAW/R16 RAW/R96P RAW/R96R LAYER_JUMP
cdrecord: Warning: DMA resid 0 for 'read buffer', actual data is too short.
cdrecord: Warning: The DMA speed test has been skipped.
Starting to write CD/DVD/BD at speed 4 in real RAW/RAW96R mode for single session.
Last chance to quit, starting real write 0 seconds. Operation starts.
cdrecord: WARNING: Drive returns wrong startsec (0) using -11634 from ATIP
cdrecord: Eingabe-/Ausgabefehler. write_g1: scsi sendcmd: no error
CDB: 2A 00 FF FF D3 5E 00 00 1A 00
status: 0x2 (CHECK CONDITION)
Sense Bytes: 70 00 05 00 00 00 00 0A 2A 30 08 80 21 02 00 00
Sense Key: 0x5 Illegal Request, Segment 0
Sense Code: 0x21 Qual 0x02 (invalid address for write) Fru 0x0
Sense flags: Blk 0 (not valid)
cmd finished after 0.002s timeout 40s
write leadin data: error after 509184 bytes
cdrecord: Could not write Lead-in.

If you let the recorder continue writing (mybe dvd) after cdrecord breakes with this error, a dvd+r gets written pretty fine, dvd-r, cd-r not ???

business_kid · 01-22-2012, 10:16 AM

First, check the small print. Does the drive *do* dvd-r?

dvd+r and dvd-r are 2 different formats structurally different. There could be separate drives for them but usually not. There _are_ separate dvds for them. There are also a whole host of clever options to allow you to get over nearly everything. on dvd-r -dummy is a dry run option.

bastl · 01-23-2012, 04:01 PM

Quote:

Originally Posted by business_kid

First, check the small print. Does the drive *do* dvd-r?

Yes, (CD/DVD+-R/+-RW/DVD-RAM), but I did get only a dvd+r to burn at a maximum speed of less than 1x.

DVD-RAM works excellent.

What are the names of the drivers/modules cdrecord wants to use? Maybe I don't compiled them in the kernel or they aren't loaded.

But it also seams that cdrecord wants to controll the writing the same way this dvd-recorder wants controll the writing , too.
Maybe this recorder has a version of cdrecord in its embedded controller ??

Can I set cdrecord to only set the recorder into write mode and sent then the/a datastream to it via DMA with no care of what the recorder does, except errors of course?

teckk · 01-23-2012, 06:33 PM

My 2 cents.

For a DVD you may try

Code:

growisofs -dvd-compat -speed=8  -use-the-force-luke=tty -Z /dev/sr0=/path/to/img.iso

For a CD

Code:

wodim  speed=8   -tao dev=/dev/sr0 /path/to/img.iso

Code:

cdrecord dev=6,0,0 speed=4 /path/to/img.iso

From the man page

Code:

-overburn
Allow cdrecord to write more than the official size of a medium. This feature is usually called overburning and depends on the fact that most blank media may hold more space than the official size. As the official size of the lead-out area on the disk is 90 seconds (6750 sectors) and a disk usually works if there are at least 150 sectors of lead out, all media may be overburned by at least 88 seconds (6600 sectors). Most CD recorders only do overburning in SAO or RAW mode. Known exceptions are TEAC CD-R50S, TEAC CD-R55S and the Panasonic CW-7502. Some drives do now allow to overburn as much as you might like and limit the size of a CD to e.g. 76 minutes. This problem may be circumvented by writing the CD in RAW mode because this way the drive has no chance to find the size before starting to burn. There is no guarantee that your drive supports overburning at all. Make a test to check if your drive implements the feature.

Code:

burnfree
Turn the support for Buffer Underrun Free writing on. This only works for drives that support Buffer Underrun Free technology. This may be called: Sanyo BURN-Proof, Ricoh Just-Link, Yamaha Lossless-Link or similar.
The default is to turn BURN-Free off, regardless of the defaults of the drive.

business_kid · 01-24-2012, 01:28 AM

So
being incorrigible - not much sympathy.

Writing dvd+r @1x: usually the medium.
control - the software actually has control but that doesn't mean the drive likes it. It's a funny relationship. They are both trying to preserve unecessary dud disks.

bastl · 01-24-2012, 12:21 PM

Quote:

Originally Posted by business_kid

So
being incorrigible - not much sympathy.

Writing dvd+r @1x: usually the medium.
control - the software actually has control but that doesn't mean the drive likes it.
It's a funny relationship. They are both trying to preserve unecessary dud disks.

I only want to burn a simple CD !!!
Two years ago it worked all fine with k3b.
I thought it is the drive and bought this one (with DVD+-R DL support, both from LG -Super Multi writer) - NO
The mediums are from intenso the same when it did yet work.
Now I think I have to study how the writing of CDs work to get behind whats going wrong or
do you have any other hints?

business_kid · 01-25-2012, 12:37 PM

Have you tried DVDRTools

It's a fork from cdrecord with a standard make system and all the extra code junked, along with hysterical warnings on running linux: "This is dvdrtools, not dvdlibc"

bastl · 01-26-2012, 01:27 PM

Yes, with dvdrtools v0.3.1 its the same error.
And I now used different mediums - always the same error.

business_kid · 01-26-2012, 03:00 PM

I am beginning to think it's the drive.

A non standard drive can get away in windows with an m$ windows driver. Seagate did this in times past - disk drivers, and you couldn't get dma on linux. Mandrake (Now mandriva) had a 'disk drake' and it had a list of dodgy drives at the time. They couldn't get dma out of seagate drives (Back in the 90s) Presume things are better now.

There are a lot of weird little options in cdrecord. If you don't want to replace the drive, read the cdrecord man page - in installments if necessary. Make sure /dev/pg0 & /dev/sg0 point to the cd drive (symlinks to /dev/whatever). Run it with -vv and see what it says. Try anything that sounds like a good idea. He does give you an accurate commentary on the drive "seems reasonable"means it's normal

bastl · 01-27-2012, 04:07 PM

Yes, but not the way that it is broken, but more that it is used wrong by the (system) tools.
So the write error could be no write error but maybe a DMA request or so ?

My dvd-recorder is SATA (/dev/sr0)

Content of file
/sys/block/sr0/device
is:

Quote:

DVDRAM GH22NS50

Because of DVD-RAM functionality the write LASER must be correct adjusted.
And reading works also with all media. So the drive must be O.K..

Maybe I compiled something in the kernel that does interacts with the writing process?

???

business_kid · 01-28-2012, 03:11 AM

dma error - I don't know.
I have slowish boxes and dma is 50-65mb/s on hard drive. If you don't have that, compile in the correct chipset driver into the kernel and get generic out - in case it nabs the chip first.

access on cd/dvd devices is usually an awful lot slower, under 10 mb/s. write is slower again. You can read, you can write, just you want it to write faster, am I correct?

BTW I can write faster, but I don't - because I have had access to a tester (Used in cdrom manufacturing). 2 clear rules came accross.
1. the faster you read a cdrom, the more mistakes are made
2. the faster you write a cdr, the more mistakes are made.

bastl · 01-29-2012, 08:40 AM

Quote:

Originally Posted by business_kid

You can read, you can write, just you want it to write faster, am I correct?

It seams that I can not write mediums-R with the -R at the end (CD-R, DVD-R, DL-R).
And yes, +R writing is too slow and also breaks with this error, but DMA continues writing,
so DVD+R gets written with less than 1x speed.

O.K I'll check my kernel.

bastl · 01-29-2012, 10:55 AM

I can not find something wrong.

business_kid · 01-30-2012, 02:14 AM

Have you experimented with cdrecord's -raw*, mmc_*, & forcespeed options?

It might be interesting to do an cdrecord <drive> -inq >file and post that.

bastl · 01-30-2012, 03:03 PM

Yes, all the same.

Quote:

cdrecord dev=6,0,0 -inq
Cdrecord-ProDVD-ProBD-Clone 3.01a03 (i686-pc-linux-gnu) Copyright (C) 1995-2010 Joerg Schilling
scsidev: '6,0,0'
scsibus: 6 target: 0 lun: 0
Linux sg driver version: 3.5.34
Using libscg version 'schily-0.9'.
Device type : Removable CD-ROM
Version : 5
Response Format: 2
Capabilities :
Vendor_info : 'HL-DT-ST'
Identifikation : 'DVDRAM GH22NS50 '
Revision : 'TN03'
Device seems to be: Generic mmc2 DVD-R/DVD-RW/DVD-RAM.

If I compile SCSI generic support into the kernel i get this error:

Quote:

cdrecord -scanbus
Cdrecord-ProDVD-ProBD-Clone 3.01a03 (i686-pc-linux-gnu) Copyright (C) 1995-2010 Joerg Schilling
cdrecord: No such file or directory. Cannot open '/dev/pg*'. Cannot open or use SCSI driver.
cdrecord: For possible targets try 'cdrecord -scanbus'.
cdrecord: For possible transport specifiers try 'cdrecord dev=help'.

So I think it is right to have it as module sg.
And if cdrecord can talk with my CD drive, like it can obviously (s.a.), it should work ???

Could there be a PAM problem? Maybe I shouldn't use PAM with cdrecord?