Bug#265747: Found it!

[Garrett McLean]

OK, I figured out exactly which alpha is causing my breakage. First I 
used debs from snapshot.debian.org.

When I used the cdrecord from cdrecord_2.0+a30.pre1-1_i386.deb 
<http://snapshot.debian.net/archive/2004/05/21/debian/pool/main/c/cdrtools/cdrecord_2.0+a30.pre1-1_i386.deb>, 
I had a successful write with no troubles. However, when I tried the 
cdrecord from cdrecord_2.0+a34-1_i386.deb 
<http://snapshot.debian.net/archive/2004/07/22/debian/pool/main/c/cdrtools/cdrecord_2.0+a34-1_i386.deb>, 
I had the same troubles that have been mentioned in this bug report.

I knew therefore that the problem must've happened somewhere in between 
a30 and 34. So, I went over to 
ftp://ftp.berlios.de/pub/cdrecord/alpha/OLD/2.01aX/ and started testing.

I installed these alphas from source, since there were no debs made for 
them.

a30: success
a31: success
a32: success
a33: failure
The process hangs, and I get stuff like this in /var/log/messages:
Feb 10 14:01:38 ludwig kernel: hdc: ATAPI reset complete
Feb 10 14:01:38 ludwig kernel: hdc: status error: status=0x08 { 
DataRequest }
Feb 10 14:01:38 ludwig kernel: hdc: status error: 
error=0x01IllegalLengthIndication
Feb 10 14:01:38 ludwig kernel: hdc: status timeout: status=0xd0 { Busy }
Feb 10 14:01:38 ludwig kernel: hdc: status timeout: 
error=0xd0LastFailedSense 0x0d
Feb 10 14:01:38 ludwig kernel: hdc: ATAPI reset complete
Feb 10 14:01:38 ludwig kernel: hdc: status error: status=0x08 { 
DataRequest }
Feb 10 14:01:38 ludwig kernel: hdc: status error: 
error=0x01IllegalLengthIndication
Feb 10 14:01:38 ludwig kernel: hdc: status timeout: status=0xd0 { Busy }
Feb 10 14:01:38 ludwig kernel: hdc: status timeout: 
error=0xd0LastFailedSense 0x0d
Feb 10 14:01:38 ludwig kernel: hdc: ATAPI reset complete
Feb 10 14:01:38 ludwig kernel: hdc: status error: status=0x08 { 
DataRequest }
Feb 10 14:01:38 ludwig kernel: hdc: status error: 
error=0x01IllegalLengthIndication

Now, I recognize that any number of things could be causing this, but I 
tested all of these alphas under the exact same conditions, and so I can 
say that something in the transition from a32 to a33 is combining with 
possible kernel and firmware bugs to make writing impossible. I don't 
know if this info should go to upstream or maybe I should just buy a new 
writer :-/

Hope it helps,
-Garrett
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]