Odd mouse issue with new xorg-server

[Tom Evans]

On Wed, May 16, 2012 at 12:51 AM, Kevin Oberman <kob6558 at gmail.com> wrote:
> On Tue, May 15, 2012 at 11:57 AM, Tom Evans <tevans.uk at googlemail.com> wrote:
>> On Thu, Apr 26, 2012 at 3:30 AM, Kevin Oberman <rkoberman at gmail.com> wrote:
>>> I hit a weird one with the new xorg-server. I have previously been
>>> running the xorg-devel from about 4or 5 months ago, so it's something
>>> that has happened since then.
>>>
>>
>> This is just a 'me too'. In my case, it seems like restarting hald
>> allows X to start up and find the mouse. Full details and logs in the
>> "Strange keyboard/hald issues after upgrading to 9-STABLE and latest
>> ports" thread.
>>
>> I see the current solutions suggest to disable hald, either by
>> completely disabling it from X, or by "AutoAddDevices off". In my
>> case, these are not solutions, as I rely on hald noticing when new
>> keyboards and mice are added to the computer and making them available
>> to X - I often dock and undock my laptop, the dock has two keyboards
>> and two mice attached to it.
>>
>> I also don't fully understand what the problem is either. Earlier
>> versions of hald/Xorg were capable of finding my mouse every single
>> time they started up. These new versions do not seem fit for use.
>
> Out of curiosity, does the mouse show up in lshal(1) output before you
> restart hald? Is there a difference after the restart?

I can check that, but I think restarting hald is not relevant. My
current approach, which seems to be working nicely, is to  boot with
gdm disabled, and restart it 3 times. This gets me a working X with a
working hal 90% of the time.

Hence I think that my idea that hald was behind this seems wrong, as
I'm now not restarting hald.

I've had enough of fiddling with this right now (I need to do some
work!), but tonight I will make the gdm rc.d script record a copy of
lshal output when it starts up, perhaps there is a time delay in hald
presenting the right devices.

Cheers

Tom