CDRW faults - OS2

This is a discussion on CDRW faults - OS2 ; Several times, the CDRW disk that I use for immediate backup of altered files has, without warning, become inaccessible to the OS. Until today I was never absolutely sure that I had not in some way caused the problem, but ...

+ Reply to Thread
Results 1 to 5 of 5

Thread: CDRW faults

  1. CDRW faults


    Several times, the CDRW disk that I use for immediate backup of altered files
    has, without warning, become inaccessible to the OS. Until today I was never
    absolutely sure that I had not in some way caused the problem, but today I am
    sure that I had no effect on the disk between a successful copy and the next
    copy which failed.

    Many tests have shown that the problem lies in the disk, and that the workaround
    is format e: /fs:udf /v /w /erase:full

    Is this as good as it gets with CDRW, or can I improve matters?




  2. Re: CDRW faults

    johnsuth@nospam.com.au wrote:

    > Many tests have shown that the problem lies in the disk, and that the
    > workaround is format e: /fs:udf /v /w /erase:full
    >
    > Is this as good as it gets with CDRW, or can I improve matters?


    Are you running the latest revision of the UDF driver? You can give a try
    deleting the CD-RW with CDRtools. Last version should be 2.01.01a04 -
    supposedly available for download on Hobbes.

    I got mixed results using DVD-RAMs and CD-RW with IBM's command-line
    software, but I never had any trouble using CDRTools and CDRecord.

    Riccardo



  3. Re: CDRW faults

    In article johnsuth@nospam.com.au writes:

    > Several times, the CDRW disk that I use for immediate backup of
    > altered files has, without warning, become inaccessible to the
    > OS. [more]
    >
    > Is this as good as it gets with CDRW, or can I improve matters?


    FWIW, I have never had much luck with CDRW disks. Maybe it has
    been the hardware I happened to use, maybe the disk brands. On
    the whole, either the disk failed to write properly and luckily
    was caught straight away or I checked after apparent success to
    find the disk unreadable on the selfsame drive. So I went over
    to CR-Rs entirely. Nowadays they're cheap enough. I have seen
    others saying similar. And, of course, USB HDDs are now common
    and cheap and have huge capacities and can be used for archival
    backups -- which is what I do very happily.
    --
    Andrew Stephenson


  4. Re: CDRW faults

    Andrew Stephenson wrote:
    >> Is this as good as it gets with CDRW, or can I improve matters?

    >
    > FWIW, I have never had much luck with CDRW disks. Maybe it has
    > been the hardware I happened to use, maybe the disk brands. On
    > the whole, either the disk failed to write properly and luckily
    > was caught straight away or I checked after apparent success to
    > find the disk unreadable on the selfsame drive. So I went over
    > to CR-Rs entirely. Nowadays they're cheap enough. I have seen
    > others saying similar.


    That's true. CDRW (and any other unprotected RW-Media too) have a
    similar sensitivity to scratches like ordinary CDs or CDRs. But they are
    very sensitive to scratches when writing. If some intensity gets lost by
    scattering, the intensity may fall below the valid range and the writing
    is no longer reliable. Because writing the ones and zero's on CDRW is
    done in one pass by modulating the laser intensity, it is even more
    sensitive. Of course, CDR share the same problem in principal. But who
    scatches an empty CDR?
    Unfortunately you won't notice the writing error unless the data fails
    to read at some point. This might not be immediately after writing,
    because the error correction may mask this ... for a while. So even if
    you would do a verify (which is uncommon in addition), you cannot be
    safe that the writing was successful.
    Sectors that are written frequently, are more likely to be affected.
    These are filesystem structures.
    With a long format the bad sectors may be remapped and the CDRW might
    come back to life for some time.

    There is only one work-around: do not use *RW media for anything but
    testing purposes.


    Marcel

  5. Re: CDRW faults

    On Wed, 7 Nov 2007 06:07:17 UTC, johnsuth@nospam.com.au wrote:

    >
    > Several times, the CDRW disk that I use for immediate backup of altered files
    > has, without warning, become inaccessible to the OS. Until today I was never
    > absolutely sure that I had not in some way caused the problem, but today I am
    > sure that I had no effect on the disk between a successful copy and the next
    > copy which failed.
    >
    > Many tests have shown that the problem lies in the disk, and that the workaround
    > is format e: /fs:udf /v /w /erase:full
    >
    > Is this as good as it gets with CDRW, or can I improve matters?


    My experience with RW media has been good, except when I use UDF with
    them. The problem appears to be, that because of the limited life of
    RW media, they just wear out a LOT faster than you would expect, when
    you use UDF. For example: If the life of the media is 1000 write
    operations, and you happen to write 1000 files to it, it appears that
    the file directory area gets written once for each file. Bingo, you
    have just written that one spot, 1000 times. The media is at the end
    of life, and you have a problem trying to write the next file (it
    isn't quite that way, since the 1000 is really only a guideline, but
    that is the general idea). I have used UDF for backup files, where I
    write one, large, file, once a week, and the media life seems to be
    fine. It is when I tried to simply copy a load of files to UDF, where
    I have run into short life situations. My solution, is to write what I
    want to a folder, then use DVDDAO to do a normal CDRW format write to
    the CD, or DVD. That writes a whole load of files, including the
    directory structure, in one pass, leaving me 999 more opportunities to
    do it again. Not quite as convenient as using UDF, but the results are
    better.

    Hope this helps...
    --
    From the eComStation 2.0 RC2 of Doug Bissett
    dougb007 at telus dot net
    (Please make the obvious changes, to e-mail me)


+ Reply to Thread