Re: Partial success and strange partial failure -was [Re: Physically have install DVD set. Want ISO's from which they came.]
On 2 March 2014 20:41, Andrei POPESCU <andreimpopescu@gmail.com> wrote:
> On Sb, 01 mar 14, 14:38:14, Richard Owlett wrote:
>>
>> root@minimal-squeeze:/home/richard# isoinfo -d -i /dev/cdrom
> ...
>> Volume size is: 1939608
>> root@minimal-squeeze:/home/richard# dd if=/dev/cdrom bs=2048
>> count=1939608 of=/home/richard/myiso1.iso conv=notrunc
>> 1939608+0 records in
>> 1939608+0 records out
>> 3972317184 bytes (4.0 GB) copied, 711.022 s, 5.6 MB/s
>
> Ok. Did you try to md5sum the image file? You can compare that to the
> md5sum of the DVD.
>
>> root@minimal-squeeze:/home/richard# dd if=/home/richard/myiso1.iso
>> bs=2048 count=1939608 of=/dev/sdb conv=notrunc
>> 1939608+0 records in
>> 1939608+0 records out
>> 3972317184 bytes (4.0 GB) copied, 1591.35 s, 2.5 MB/s
>
> Not sure if this makes a differences, but why the count= and conv= ?
Richard is using count and conv because I pointed him to the instructions here:
http://www.debian.org/doc/manuals/debian-reference/ch09.en.html#_making_the_iso9660_image_file
Those instructions require the count and notrunc, with this warning:
"You must carefully avoid ISO9660 filesystem read ahead bug of Linux
as above to get the right result."
The above is Richard showing that he complied with those instructions.
I would like to have more information about this "read ahead bug", but
I cannot find a full bug report or similar technical discussion of it.
Thomas states that "Debian i386 images are ISO 9660 filesystems" so I
assume that this bug workaround is still required for them.
The best information I can find is this recent post by Thomas.
http://askubuntu.com/questions/408224/the-file-casper-filesystem-squashfs-is-corrupt/410492#410492
Thomas thank you for the information you already posted, I found it
very interesting. Can you recommend anywhere I can read to learn more
about the Debian DVD filesystems? Apart from searching for your past
emails and forum posts :)
And where can I learn more about the read-ahead bug: how to identify
it, how to avoid it, is there a bug tracker for it, which if any
source file will eventually contain a fix for it? Does it also affect
udf file systems?
Reply to: