Re: dvd writes truncated 3 Mbytes

[Gene Heskett]

On Saturday 19 November 2005 05:16, address@hidden wrote:
>Hi,
>
>> K3B reports it has written 1160 of 1163 Mbytes each time, but doesn't
>> seem to have a problem with that.
>> I just ran growisofs from the cli, and it also gets exactly the same
>> bad md5sum, not reporting the final write.  Here is the cli line &
>> end of burn session reports:
>> builtin_dd: 595536*2KB out @ average 3.9x1385KBps
>
>Well, 595536*2*1024 yields 1219657728 = 1163.15625 GB.
>At least growisofs seems to have seen 1163 GB of input.
>Maybe a short count is normal with K3B ? (I got no KDE)
>
>> address@hidden FC4]# md5sum </dev/cdrom
>> 292969c8f81e7c0e3ed0ee76a5a637ad  -
>> This last is the same bad md5sum I've seen 4 times before.
>
>Looks like quite a constant behavior.
>Not very typical for writer problems.

That was my thought too.

>Did you check how many bytes are actually read from the media ?
>  wc </dev/cdrom

Thats running now.  Returns:
address@hidden FC4]# wc </dev/cdrom
4427975 26174256 4700372992
Which is odd indeed as the last figure, supposedly the bytes, is the
disks capacity, nowhere near used up.

>If this counts less than 1219657728 bytes then the image is
>indeed incomplete.
>/dev/cdrom may well yield more bytes than you have written to media.
>(Especially with re-used RW media. What type are you using ?)

Staples branded 2.5x +RW's, and a spindle of TDK 4x +RW's.  Each disk
was new, but both have been reburnt at least 2 times now.

>For MD5 comparison with the ISO file on hard disk you will have
>to truncate the stream from /dev/cdrom to that file's exact size.
>Maybe a run of program diff brings some enlightenment.

After perusing the man page for diff, I have a diff -e running now.
Which, like the optionless invocation, only returns that they differ,
but not at what byte..So now I'm running a cmp -l on them.

And that returns:

address@hidden FC4]# cmp -l CAELinuxBeta1.iso /dev/cdrom
cmp: EOF on CAELinuxBeta1.iso

So there is no difference up to the EOF of the src iso!

>> Has anyone else encountered a similar problem?
>
>Not yet. But none of my rewritable media would deliver
>the correct checksum with untruncated input from /dev/cdrom.

Which is obviously the problem, but what to do to fix that on a
permanent basis?  There needs to be some sort of a way to tell proggies
such as md5sum that byte so and so is the last byte to read & that it
should stop and print the sum as it exists at that point to stdout.

>> I've been making good cd's all along with no problems, in
>> this new drive, till now.
>
>Something must have changed.
>A different size of ISO image ?

First time in 2-3 years I've tried to burn a dvd iso.  My earlier
attemps with a previous drive just made coasters so I shelved the idea
till the software matured.  I'd originally bought the drive and a
spindle of disks intending to use them with amanda, but at 4.7
decimal gigs, I was just trading a poor tape format in on an equally
limited in reliabilty disk.  Since hard drives have gotten pretty
dependable, I finally bought a 200Gb drive and set amanda up to use
virtual disks on it.  Best move I ever made.

So, based on that, I think I'll add address@hidden to the
address line and see if they have any suggestions regarding this
apparent inability to use md5sum against a dvd, whereas it works
flawlessly for a cd in the same drive.

>Have a nice day :)

You've improved it considerably by pointing out that all 5 burns I've
made so far, are indeed correct, but we have no ready means to test
when the media is a dvd +RW disk.

>Thomas

-- 
Cheers, Gene
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
99.36% setiathome rank, not too shabby for a WV hillbilly
Yahoo.com and AOL/TW attorneys please note, additions to the above
message by Gene Heskett are:
Copyright 2005 by Maurice Eugene Heskett, all rights reserved.