Re: [Halevt-dev] Using parent values in exec attribute

[Patrice Dumas]

On Tue, Mar 23, 2010 at 01:20:58PM +0100, Marcus Habermehl wrote:
> 
> hal_interface.c:480 (halevt_run_command) Running: notify-send -i
> gnome-dev-printer "USB Printer attached" "USB Printer Brother
> Industries, Ltd DCP-195C was attached."
> 
> lshal -m on removal the printer:
> 
> 13:12:32.905: usb_device_4f9_222_BROM9F131604_if0 removed
> 13:12:32.906: usb_device_4f9_222_BROM9F131604_if1 removed
> 13:12:32.910: 
> usb_device_4f9_222_BROM9F131604_if2_scsi_host_0_scsi_device_lun0_scsi_generic
> removed
> 13:12:32.912: storage_serial_Brother_DCP_195C_BROM9F131604_0_0 removed
> 13:12:32.912: usb_device_4f9_222_BROM9F131604_if2_scsi_host removed
> 13:12:32.914:
> usb_device_4f9_222_BROM9F131604_if2_scsi_host_0_scsi_device_lun0
> removed
> 13:12:32.915: usb_device_4f9_222_BROM9F131604_if2_scsi_host_0 removed
> 13:12:32.915: usb_device_4f9_222_BROM9F131604_if2 removed
> 13:12:32.930: usb_device_4f9_222_BROM9F131604 removed

Unless I am missing something, it looks like the parent is 
removed after the device. So if things are correctly serialized,
it should work correctly. I would like to rule out an 'incorrect'
order of callbacks to halevt before investigating a bug in halevt.

I just committed a change to ease debugging. Could you please
update and run halevt with the -d option? It will stay in the 
foreground and print some information, could you please redirect 
stderr to a file, insert and remove the device and post the result?

Thanks.

--
Pat