Re: unable to read some of a DVD contents

2005-05-16 Thread scdbackup
Hi,

 As you mentioned, neither the burner nor the media 
 is the one to blame so what
 is the thing that causes the test checksum to fail?

In this case the individual media is to blame.
At least in conjunction with my burner device.

The annoying problem with DVD burning is that it
is so hard to predict wether an individual
combination of burner and media will work
flawlessly. So, in case of trouble, one will most
often get the advice to either use other media or
upgrade the drive's firmware.
Of course, this is the expression of pure
helplessness. We are poking in the mist by
help of very long sticks.


 Does that cause by the on-the-fly burn using growisofs?

It should not. (But who knows for sure ?)
Modern burners are protected against buffer
underrun which used to be the main problem
of burning on-the-fly with old CD recorders.

Up to now i could not find a correlation
between obvious slowdowns of the feeding
data stream and failure to checkread the
resulting media.


Have a nice day :)

Thomas


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: Re: unable to read some of a DVD contents

2005-05-16 Thread Kadir Wijaya
Yes. as Thomas and Volker pointed that the media is the one to be blamed.

I used better quality DVD, the most expensive one in the store :), and I've 5 
successful burns and all passed the integrity check.

Thanks for all your helps,
Kadir
---
This email contains confidential information. If you have received this
email in error, please delete it immediately, and inform us of the mistake
by return email. Any form of reproduction, or further dissemination of
this email is strictly prohibited.
Also, note that the opinions expressed in this email are those of the 
author and are not necessarily those of the Fonterra Co-operative Group
Limited.
http://www.fonterra.com/
---



Re: unable to read some of a DVD contents

2005-05-15 Thread Kadir Wijaya
I just had to trash a previously used DVD-RW which failed to pass the checksum 
test with errors at the very start of the image. 
After a test re-write it failed to deliver the last 2 MB (but the start of the 
test image was ok). 
A third re-write yielded missing 300 MB at the end of the media. The fourth 
try failed with :-[ PERFORM OPC failed with SK=3h/ASC=73h/ACQ=03h]: 
Input/output error According to keys.txt : 3 73 03 POWER CALIBRATION AREA 
ERROR (Not much idea what that means but it looks ugly and fatal) 
I successfully tested another DVD-RW from the same spindle as the ill one. So 
the burner is not to blame. 
It is not about the manufacturer of the media, not even about the particular 
lot. It is not about the number of re-uses. 
The sucessfull DVD-RW is in heavy test use, has a nice collection of scratches 
... and it works fine. The ill one had its third or fourth re-use, looks shiny 
and smooth ...
As you mentioned, neither the burner nor the media is the one to blame so what 
is the thing that causes the test checksum to fail?
Does that cause by the on-the-fly burn using growisofs?

Thanks,
Kadir

 

---
This email contains confidential information. If you have received this
email in error, please delete it immediately, and inform us of the mistake
by return email. Any form of reproduction, or further dissemination of
this email is strictly prohibited.
Also, note that the opinions expressed in this email are those of the 
author and are not necessarily those of the Fonterra Co-operative Group
Limited.
http://www.fonterra.com/
---



Re: unable to read some of a DVD contents

2005-05-15 Thread Volker Kuhlmann
 The sucessfull DVD-RW is in heavy test use, has a nice collection of
scratches ... and it works fine. The ill one had its third or fourth
re-use, looks shiny and smooth ...

 As you mentioned, neither the burner nor the media is the one to blame
 so what is the thing that causes the test checksum to fail?

Yes the media *is* to blame. You always get some faulty ones more or
less randomly (that's why he said it wasn't the whole lot which was
shot).

 Does that
 cause by the on-the-fly burn using growisofs?

Definitely not.

Volker

-- 
Volker Kuhlmann is possibly list0570 with the domain in header
http://volker.dnsalias.net/ Please do not CC list postings to me.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: Re: unable to read some of a DVD contents

2005-05-13 Thread Volker Kuhlmann
 I used verbose option and clearly see that the burn process was completed and 
 the session was closed.
 what is the indicator that the burn process was only 95% completed?

The burn was 100% completed at 95% quality. ;)

That's putting it colloquially of course. The burner is maladjusted to
the media, when all remaining tolerance for successful reading is
exceeded, you get a read error. What this really indicates however that
the whole process is very marginal with no space for future errors left.
Dump those disks, or at least don't expect them to still hold your
precious data in a few months. And even if they still work in a few
months, don't trust it.

Volker

-- 
Volker Kuhlmann is possibly list0570 with the domain in header
http://volker.dnsalias.net/ Please do not CC list postings to me.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: Re: unable to read some of a DVD contents

2005-05-13 Thread scdbackup
Hi,

 what is the indicator that the burn process was only 95% completed?

The only reliable indicator seems to be checkreading
by all drives which shall be able to read the DVD.

This would mean either

- to compare each file on the DVD with its original
  on hard disk (if the files on disk did not change
  meanwhile)

- or to compare the data image on DVD with the data
  image on disk (if the image was stored on disk)

- or to compare the data image on DVD with a checksum
  which was made when that image was written to DVD.
  (I prefer this method.)


I just had to trash a previously used DVD-RW which
failed to pass the checksum test with errors at the
very start of the image. After a test re-write it
failed to deliver the last 2 MB (but the start of the
test image was ok). A third re-write yielded missing
300 MB at the end of the media.
The fourth try failed with 
   :-[ PERFORM OPC failed with SK=3h/ASC=73h/ACQ=03h]: Input/output error
According to  keys.txt  : 
   3  73  03   POWER CALIBRATION AREA ERROR
(Not much idea what that means but it looks ugly and fatal) 

I successfully tested another DVD-RW from the same
spindle as the ill one. So the burner is not to blame.

It is not about the manufacturer of the media, not even
about the particular lot. It is not about the number of
re-uses. The sucessfull DVD-RW is in heavy test use, has
a nice collection of scratches ... and it works fine.
The ill one had its third or fourth re-use, looks shiny
and smooth ...

A theory which emphasizes the influence of moon phases or
geomantic radiation would probably be as successful as
the usual methods of DVD compatibilty prediction.


Have a nice day :)

Thomas


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: unable to read some of a DVD contents

2005-05-13 Thread The Wijaya
Thanks for the comprehensive explanation,

 Hi,

  what is the indicator that the burn process was only 95% completed?

 The only reliable indicator seems to be checkreading
 by all drives which shall be able to read the DVD.

 This would mean either

 - to compare each file on the DVD with its original
   on hard disk (if the files on disk did not change
   meanwhile)

This is the method we used (comparing each file on the DVD with its orginal
on hard drive by using 'cmp'  command)


 - or to compare the data image on DVD with the data
   image on disk (if the image was stored on disk)

 - or to compare the data image on DVD with a checksum
   which was made when that image was written to DVD.
   (I prefer this method.)

First time heard about this method. How to actually obtain the DVD checksum?
Will this method run faster than the first one? (We normally have around
1 files to compare).
Brief explanation on this method will be appreciated (keen to try this out).



 I just had to trash a previously used DVD-RW which
 failed to pass the checksum test with errors at the
 very start of the image. After a test re-write it
 failed to deliver the last 2 MB (but the start of the
 test image was ok). A third re-write yielded missing
 300 MB at the end of the media.

How did you know there was 300MB data missing at the end of media?
In our case, the file sizes exactly match the file sizes on the harddrive.


 The fourth try failed with
:-[ PERFORM OPC failed with SK=3h/ASC=73h/ACQ=03h]: Input/output error
 According to  keys.txt  :
3  73  03   POWER CALIBRATION AREA ERROR
 (Not much idea what that means but it looks ugly and fatal)


Got this problem before upgrading growisofs.

 I successfully tested another DVD-RW from the same
 spindle as the ill one. So the burner is not to blame.

 It is not about the manufacturer of the media, not even
 about the particular lot. It is not about the number of
 re-uses. The sucessfull DVD-RW is in heavy test use, has
 a nice collection of scratches ... and it works fine.
 The ill one had its third or fourth re-use, looks shiny
 and smooth ...


 Have a nice day :)

 Thomas


Thanks,

Kadir


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: unable to read some of a DVD contents

2005-05-13 Thread scdbackup
Hi,

 - or to compare the data image on DVD with a checksum
which was made when that image was written to DVD.
 
 First time heard about this method. How to actually obtain the DVD checksum?

If you got the image stored on disk you may use 
program md5sum.
If you generate and burn the image on-the-fly
compute the checksum by a pipe filter. With bash
there is a trick called Process Substitution
which would allow to do that by md5sum, too.

On Aug 25 2004, Robert Schiele wrote regarding
a related topic :
:  On bash you can do simply:
:  $ mkisofs ... | tee (md5sum) | cdrecord

Ok, it's not that simple but it shows a path.

You will have to record the output of md5sum
as well as the exact size of the image. The
size is important since you will hardly get
an exact EOF from DVD or CD. At least with
CD you need to pad about 300 kB after the
the checksumed image to ensure that the
image can be completely read. (DVD unsure)

So you would need a script which invokes
md5sum, counts the bytes and records them in
some kind of database. Best is if the checksum
gets appended to the image on DVD, too.

In my own project i use a C language program
for that. So i do not have to rely on exotic
(...) gestures. This also allows additional
stunts. For details see
 http://scdbackup.webframe.org/README
search for Verifying CDs and DVDs and
Appendix VERIFY. See also 
 http://scdbackup.webframe.org/cd_backup_planer_help
for details about some options mentioned in
README.


 Will this method run faster than the first one? 
 (We normally have around 1 files to compare).

It should not be slower. It avoids directory lookups
and random access movements. On my system reading
begins with about 4.8 MB/s and accelerates up to
10 MB/s while computing uses up 30% of an AMD 2600+.
A full DVD verify of 4.7e bytes lasts about
10 minutes if the media is ok. Damaged media are
often quite slow.


 Brief explanation on this method will be appreciated (keen to try this out).

To make it brief would be serious work ;-)


 How did you know there was 300MB data missing at the end of media?
 In our case, the file sizes exactly match the file sizes on the harddrive.

Those sizes are read from the ISO filesystem 
directory which is located at the beginning
of the disk. That part of your DVD seems to 
be ok. But it refers to data areas which seem
to be damaged.

The missing size was detectable because my
checksum records contain the size measurement
from writing.

In my role as developer of backup software
i am looking for trouble. Therefore i not only
compute a single checksum for the whole backup
but one for each 64 kB of data. So i can
recognize wether the damage forms single spots
or wether it hits wide areas.
(For details see README, Appendix REDUNDANCY)


 :-[ PERFORM OPC failed with SK=3h/ASC=73h/ACQ=03h]: Input/output error
 Got this problem before upgrading growisofs.

Well, i'm on 5.21. No newer version known.
Also, the problem is restricted to a single disk.
I'll try cdrecord-ProDVD on it next.

It's just for illustration how important a
trustworthy checkread is.


Have a nice day :)

Thomas


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



unable to read some of a DVD contents

2005-05-12 Thread Kadir Wijaya
Hi there,

I've successfully burnt data (4 GB, around 1 files) to a DVD. 
I saw that all the files have been burnt to DVD with exact size.
However, I couldn't read some of the files (corrupted ?) and this happened 
randomly (i.e. I burnt another ten DVDs with the same contents and the 
corruptions happened on different files).

Here is the error message while copying the file (linux):
cd:reading /media/cdrecorder1/ares_apca_pref_050520202.zip Buffer I/O error.
 
and the message log output the following messages:
May 13 12:10:59  kernel: scsi6: ERROR on channel 0, id 0, lun 0, CDB: Read (10) 
00 00 05 71 15 00 00 23 00
May 13 12:10:59  kernel: Current sr3: sense key Medium Error
May 13 12:10:59  kernel: Additional sense: Unrecovered read error
May 13 12:10:59  kernel: end_request: I/O error, dev sr3, sector 1426516
May 13 12:10:59  kernel: Buffer I/O error on device sr3, logical block 356629

If I copy the file in a windows box it gives me a cyclic redundancy error 
message.

Below is the configuration of software and OS used:
1. OS  Linux 2.6.9-5.0.5.ELsmp 
2. growisofs -  version 5.21
3. mkisofs 2.01 (i686-pc-linux-gnu)
4. burner - SONY DRU-700A (with latest firmware upgrade)

Following is the growisofs command used:
/usr/local/bin/growisofs -use-the-force-luke -v -iso-level 2 -pad -Z $g_device 
-R -J $buffer_path

I wonder if anyone out there can point me out what's going wrong here.

Thanks,
Kadir

 

---
This email contains confidential information. If you have received this
email in error, please delete it immediately, and inform us of the mistake
by return email. Any form of reproduction, or further dissemination of
this email is strictly prohibited.
Also, note that the opinions expressed in this email are those of the 
author and are not necessarily those of the Fonterra Co-operative Group
Limited.
http://www.fonterra.com/
---