Google
  Web www.spinics.net

Re: : Again on: How to check *CDROM* correctness?

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


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Sun, 6 Apr 2003 07:26:44 +0200, M. Fioretti wrote:

> Thanks for all your input. Summing everything up, the conclusion seems
> to be that, when you have no original ISO, but only the *CDROM* made
> out of them God knows how, there are only two options:
> 
> 	1) Boot with those CDs, run linux mediacheck on all of them,
>            and compare the results with those published on the
> 	   original site
> 
> 	2) (in some cases only) mount the CD in linux and run md5sum
> 	   on the corresponding device:
> 
> 		mount -t iso9660 /dev/cdrom /mnt/cdrom
> 		md5sum /dev/cdrom

Mounting the CD is not necessary for "md5sum /dev/cdrom" to work.

> For the record, I have tried 2) on the CDROM I used to create
> *this* RH 8.0 I'm writing from, and the result is:
> 
> 	[root@xxxxxxx root]# md5sum /dev/cdrom
> 	md5sum: /dev/cdrom: Input/output error
> 
> 	[root@xxxxxxx root]# cat /dev/cdrom | md5sum
> 	cat: /dev/cdrom: Input/output error
> 	but a checksum was indeed output
>   
> At this point, I'll probably do both as soon as I get those shrike CDs from my
> friend, but any general comment is still welcome, of course.

This is due to run-out sectors at the end of the CD. It puts more
sectors than desired into the MD5 checksum. 

You can try the following: Look up the size of the original ISO
images. Divide the size by 2048 and then use "dd" to pipe the CD
contents to "md5sum". This may or may not get rid of the I/O
errors. 

Alternatively, try reading the CD with the "readcd" commmand's
function "11". There you can also specify the number of blocks to
read. The resulting ISO images should have exactly the same size
than the original ISO images.

However, if you get I/O errors in either case, the CDs should have
been burnt in DAO mode or TAO mode _with_ padding. In case of the
former, you would not have any problems. In case of the latter,
you could cut sectors at the end without trouble.

- -- 
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)

iD8DBQE+kCQ00iMVcrivHFQRAowYAJ9sMaZmqVsjfikNuTGPdPNOOV6JMgCfUbnJ
I2yUI2EGzQFNuhRL5dqGbH4=
=re++
-----END PGP SIGNATURE-----





[Fedora Users]     [Centos Users]     [Kernel Development]     [Red Hat Install]     [Red Hat Watch]     [Red Hat Development]     [Red Hat Phoebe Beta]     [Yosemite Forum]     [Fedora Discussion]     [Gimp]     [Stuff]     [Yosemite News]

Powered by Linux