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

Re: Checking the md5sum of a burned CD



Paul E Condon wrote :
> On 2009-03-19_05:29:17, M. Lewis wrote:
>> thveillon.debian wrote:
>>> M. Lewis a ??crit :
>>>> I've burned a CD from a .iso image. Now I want to verify the burning
>>>> process worked correctly (there were no errors reported). Note, I
>>>> verified the md5sums of the .iso's before I burned them. So now I'd like
>>>> to check the burned CD against that md5sum.
>>>>
>>>> Thanks,
>>>> Mike
>>> Hi,
>>>
>>> "md5sum /dev/cdrom" or "md5sum /dev/scd0" works here (Squeeze and
>>> Lenny), whatever applies to your system.
>>> If you have I/O error check that it's not mounted, and not accessed by
>>> some other process (like "kio*" in KDE).
>>>
>>> Tom
>> Great. Thanks Tom. I was totally unaware it could be done as simply as that.
>>
> 
> Several months ago, when Lenny was still 'testing', I had difficulty
> with this: On my hardware, the read from /dev/cdrom had extra bytes at
> the end and was a longer byte stream than the .iso image from which
> the CD was burned. So, of course, compare of md5 sums between CD and
> .iso failed. There is a work-around in that one can use dd to copy
> from /dev/cdrom just enough blocks to match the length of the .iso
> file, and then compute the md5sum of that (truncated) length matching
> file. OTOH the problem may not have been hardware, and it may have
> been fixed. So if at first you get results that indicate that you have
> a badly burned CD, it may be the test that is bad, not the burn.
> 
> I have not verified this situation recently, so this is just a warning.
> I think there are many people like me who know about this old problem,
> so it would be nice for you to report back to the list as to whether or
> not it is still there.
> 

Following a similar thread on the French list where this problem was
also related, I did some testing with dvd's and cd's, GNU-Linux distros
and movies burnt with K3B, on Lenny "almost" stable and lately on
Squeeze. I didn't come across any issue, so at least here it works fine.

Tom


Reply to: