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

RE: Fwd: FW: [Bug 47663] New: DVD burned from iso file has incorrect checksum



Hi Andy,

In the meantime I tried to write an iso-image ( FC3-i386-DVD.iso - fedora core 3 distribution ) to a DVD+R.
I used the same parameters as created by k3b and added the -dvd-compat option.

Also here is a problem concerning the size:
the original dvd has 602151 4k-blocks, while the DVD after burning has one additional 4k-block.

Did I miss something, or is there a bug in growisofs ?

Have a nice weekend,

Winfrid

 
Fujitsu Siemens Computers S CE DE PCO Süd

Winfrid Tschiedel

Domagkstrasse 28

80807 München

Tel:: ++49-89-3222-1532

FAX: ++49-89-3222-329-1532

intranet : Winfrid.Tschiedel@fujitsu-siemens.com

internet: Winfrid.Tschiedel@hpc.fujitsu-siemens.com



-----Original Message-----
From: Andy Polyakov [
mailto:appro@fy.chalmers.se]
Sent: Wednesday, November 03, 2004 5:56 PM
To: Tschiedel, Winfrid
Cc: cdwrite@other.debian.org
Subject: Re: Fwd: FW: [Bug 47663] New: DVD burned from iso file has incorrect checksum

> Many thanks for your explanation, nevertheless, currently there 2
> things which I cannot test for the moment, is it true, that you are
> saying :
>
> using DVD-RW would report a correct checksum ( here I know, that at
> least there is one burning program under WINDOWS, which generates DVDs
> with a correct checksum )
>
> using DVD-R or DVD+R would be also okay .

Yes. If unit didn't have to pad the file system image as required by specification, then a) DVD-RW Sequential, b) DVD-R and c) DVD+R recordings performed with growisofs -dvd-compat(!) option should pass the 'md5sum /dev/dvd' test. "Should" means "I won't be surprised if someone runs into a firmware that will fail this test even for above mentioned recordings."

In any other combination, namely a) padding to 32KB boundary was required, b) growisofs executed without -dvd-compat option, c) DVD+RW and d) DVD-RW Restricted Overwrite recordings, trivial 'md5sum /dev/dvd'
test might *fail*. The failure however won't be violation of any specification or indicate a bug in recording software or unit firmware.

As you can see it's a number of combinations and if you don't want to figure out which particular combination caused the above mentioned checksum test to fail at any particular occasion, always verify as much data as you've actually written: 'dd if=/dev/dvd count=<appropriate
count> | md5sum' and not just 'md5sum /dev/dvd'. A.


Reply to: