[EMAIL PROTECTED] (Johan Vromans) wrote:
> Inspection shows that all files on the CD are accessible and
> apparently correct. I tried "find . -type f -print | xargs md5sum".
> The problem shows only when comparing the CD with the image, e.g.,
> "dd if=/dev/hdc | cmp - image.iso". Maybe it is the "read ahead" bug?
It's not a valid thing to do AFAIK. What is being written validly is
the data saved by mkisofs, and the way to check that is (as you sort-of
did) with md5sum. I usually loopback mount the image, cd to the mount
point, save the md5 info, and then use the -c option to verify the CD.
That's easier than hand compare (which I think you did) but no more or
less valid.
In any case, the dd includes the padding after the data, and reads
until it gets an error. Testing that is not valid, since it's not part
of the data, just trailing non-data noise. If you use dd, you should use
count= and bs= to limit the CD read to the length of the image, and I
wouldn't bet that even that is meaningful.
The post data info on a burn is somewhat like reading the inter-record
gap on a floppy. It's there, it's important that it be there, but it's
not part of the data.
Is that clearer?
--
-bill davidsen ([EMAIL PROTECTED])
"The secret to procrastination is to put things off until the
last possible moment - but no longer" -me
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
