[Bug 14394] cdrecord/k3b/nautilus problem

[Ralph Corderoy]

Public bug report changed:
https://launchpad.net/malone/bugs/14394

Comment:

> I found that most kernel error messages can be eliminated by killing
> the hald-addon-storage processes (there are two).

My older Ubuntu system, 5.04, doesn't have hald-addon-storage processes.

    $ ps xaww | g hald
     6872 ?        Ss     0:01 /usr/sbin/hald --drop-privileges
    29664 pts/1    S+     0:00 grep -E hald
    $

If I strace hald it's polling several file descriptors and regularly opening
/dev/scd0 to ioctl(CDROM_DRIVE_STATUS) it.

    $ sudo strace -p 6872
    Process 6872 attached - interrupt to quit
    gettimeofday({1138625597, 428745}, NULL) = 0
    gettimeofday({1138625597, 428978}, NULL) = 0
    poll([{fd=6, events=POLLIN}, {fd=10, events=POLLIN}, {fd=8, events=POLLIN}, {fd=9, events=POLLIN}, {fd=14, events=POLLIN|POLLPRI}], 5, 849) = 0
    gettimeofday({1138625598, 279090}, NULL) = 0
    open("/dev/scd0", O_RDONLY|O_NONBLOCK|O_EXCL|O_LARGEFILE) = 3
    ioctl(3, CDROM_DRIVE_STATUS, 0x7fffffff) = 2
    close(3)                                = 0
    time(NULL)                              = 1138625598
    gettimeofday({1138625598, 306502}, NULL) = 0
    poll(...

    $ sudo strace -ttt -p 6872 2>&1 | g /dev/scd0
    1138625702.325428 open("/dev/scd0", O_RDONLY|O_NONBLOCK|O_EXCL|O_LARGEFILE) = 3
    1138625704.327099 open("/dev/scd0", O_RDONLY|O_NONBLOCK|O_EXCL|O_LARGEFILE) = 3
    1138625706.327802 open("/dev/scd0", O_RDONLY|O_NONBLOCK|O_EXCL|O_LARGEFILE) = 3
    1138625708.328481 open("/dev/scd0", O_RDONLY|O_NONBLOCK|O_EXCL|O_LARGEFILE) = 3

Perhaps you can monitor hald-addon-storage to see what it does when
recording is interrupted.

Alternatively, use lshal (search output for /dev/scd0) and
hal-[gs]et-property to tinker with some of the options and see if one of
them turns off the mucking around with /dev/scd0 during record, e.g.
auto-mount.

    $ hal-get-property --udi /org/freedesktop/Hal/devices/scsi_0_0_4_0 \
    >     --key scsi.target
    4