HAL taking over

[martinko]

Kevin Oberman wrote:
>> From: Joe Marcus Clarke <marcus at marcuscom.com>
>> Date: Thu, 02 Nov 2006 20:52:04 -0500
>>
>> On Thu, 2006-11-02 at 14:39 -0800, Kevin Oberman wrote:
>>> HAL looks wonderful, but it seems to be trying to take control of my
>>> system. 
>>>
>>> I have my system set up to ignore the lid switch
>>> (hw.acpi.lid_switch_state: NONE)
>>>
>>> When I have HALD running, it suspends my system, anyway, and it does not
>>> resume. It reboots.
>> You will have to ignore the lid button in HAL.  Basically, create
>> a /usr/local/share/hal/fdi/preprobe/20thirdparty/10-acpi-lid.fdi file
>> with the following contents then restart hald:
>>
>>   <?xml version="1.0" encoding="UTF-8"?>
>>   <deviceinfo version="0.2"> 
>>     <device> 
>>       <!-- ignore ACPI lid buttons -->
>>       <match key="button.type" string="lid">
>>         <merge key="info.ignore" type="bool">true</merge>
>>       </match>
>>     </device>
>>   </deviceinfo>
>>
>> If that still doesn't work, HAL may need a patch to support ignoring
>> these kind of events.
>>
>>> When I have HALD running, I can no longer unmount disks. I can't do it
>>> from the command line (volume busy) or from nautilus (Not authorized).
>> Which disks?  Volumes mounted by HAL can be controlled with the
>> gnome-mount command provided you have appropriate access (see
>> http://www.freebsd.org/gnome/docs/faq2.html#q19 ).
>>
>>> I suspect HALD can be configured to make all of this work right, but I
>>> don't know quite where to look. (OK, I have some hints on the unmounting
>>> issue, but I have not gotten it to work yet.
>>>
>>> Finally, the disks are no longer labeled in a meaningful fashion. What
>>> used to be the "D" drive on my system is now showing up as "7.8 GB". Not
>>> too useful and made a bit worse because I often mount two partitions of
>>> almost exactly the same size.
>> You can label volumes using tunefs -L.  After doing this, they should
>> appear with more meaningful names.
> 
> Mixed success. First, The big one...the XML for the lid worked
> perfectly. I can now close the lid without suspending!
> 

Shouldn't HAL after installation respect the current system behaviour?
(that is e.g. to install above mentioned script if
hw.acpi.lid_switch_state is set to NONE)

M.