Odd floppy disk error

From: Kimmo Sundqvist (rabbit80_at_mbnet.fi)
Date: 09/27/03

  • Next message: David Ford: "LInksys WMP11 (BCM4301 chip)"
    To: linux-kernel@vger.kernel.org
    Date: 	Sun, 28 Sep 2003 00:13:04 +0300
    
    

    Hello

    I was testing old 1.44MB diskettes with badblocks, like "badblocks -sv
    /dev/fd0" and this one particular diskette passed the test twice. I then
    decided to look into a text file that resided on the diskette.

    The floppy drive started reading. It seemed to take longer than expected, so
    I forgot it and went to do other things. After some hours had passed, I
    wondered why the load on the machine was 100% in both CPUs all the time.

    Running top told me that
       7 5 -10 0 0 0 0 0 1 SW< 85.1 0.0 12:28 events/1
      218 25 0 38 2244 1388 856 0 0 R 60.0 0.1 35:54 klogd
      215 15 0 70 1420 588 832 0 0 R 36.2 0.0 26:33 syslogd

    and I quickly figured out that my /var partition was full.

    Filesystem 1K-blocks Used Available Use% Mounted on
    /dev/hda7 921748 921732 16 100% /var
            
    Looking for explanations, the syslog was

    -rw-r----- 1 root adm 212582400 Sep 27 22:44 syslog

    and the contents of it were mostly

    Sep 27 22:44:04 minjami kernel: end_request: I/O error, dev fd0, sector 1327
    Sep 27 22:44:04 minjami kernel: Buffer I/O error on device fd0, logical block
    1327
    Sep 27 22:44:04 minjami kernel: floppy0: disk absent or changed during
    operation

    repeated for sectors from 1327 to 1334 in sequence, over and over again.

    What puzzles me is, why is the message there 2050 times in one second? That
    makes 6150 lines in one second. It begins around line 850 000, and syslog is
    around 2 245 000 long. Timespan of 3 minutes, 46 seconds. Therefore it's
    6150 lines per second throughout.

    You sure have made a fast floppy driver, congratulations. But the actual
    point in writing this is that if there's anything unintentional in this
    behaviour, I just wanted to let you know.

    "Linux minjami 2.6.0-test4-mm2 #1 SMP Thu Aug 28 15:08:11 EEST 2003 i686
    GNU/Linux" is a mixed Debian 3.0r1 stable/testing/unstable system.

    /var/log/messages had only these two lines repeated (equal amount of times is
    my guess)

    Sep 27 22:44:07 minjami kernel: end_request: I/O error, dev fd0, sector 1334
    Sep 27 22:44:07 minjami kernel: floppy0: disk absent or changed during
    operation

    and the contents of /var/log/kern.log were identical to the contents of
    /var/log/syslog between 22:40:14 and 22:44:07

    For more information, mail me, since I'm not subscribed.

    -Kimmo Sundqvist

    While writing this I gently removed the floppy from the drive, but the busy
    light has been glowing for minutes after it now, and the load is still up.
    The system is 100% responsive though.

    -
    To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
    the body of a message to majordomo@vger.kernel.org
    More majordomo info at http://vger.kernel.org/majordomo-info.html
    Please read the FAQ at http://www.tux.org/lkml/


  • Next message: David Ford: "LInksys WMP11 (BCM4301 chip)"

    Relevant Pages

    • Odd floppy disk error
      ... Sep 27 22:44:04 minjami kernel: Buffer I/O error on device fd0, ... You sure have made a fast floppy driver, ... GNU/Linux" is a mixed Debian 3.0r1 stable/testing/unstable system. ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: udev sysfs docs Re: State of devfs in 2.6?
      ... floppy is always present in the system, but I access it like once a month ... driver non-stop in unswappable kernel memory is wastefull. ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: PROBLEM: floppy motor spins when floppy module not installed
      ... >> in the BIOS settings so the hard disk is tried before floppy. ... >> ever need to boot from a floppy, ... This seems like the correct solution to me. ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: floppy regression from "[PATCH] fix floppy.c to store correct ..."
      ... > causes an annoying side-effect. ... Upon first write attempt to a floppy I get ... On later opens, we have done, so ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • floppy regression from "[PATCH] fix floppy.c to store correct ..."
      ... causes an annoying side-effect. ... Upon first write attempt to a floppy I get ... dd: writing `/dev/fd0': Operation not permitted ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)