Re: Stopping modemmanager from auto probing?

[Dan Williams <dcbw redhat com>]

On Tue, 2010-03-09 at 11:16 +0100, Torsten Spindler wrote:
> Hello,
> 
> Alex and I have submitted a patch for the blacklisting in ModemManager
> in this bug report:
> https://bugzilla.gnome.org/show_bug.cgi?id=608022
> 
> If you have some time, can you please let us know if it can be included.
> Is there anything else we can do to move this forward?

Not until I look at it; will get there soon.

Thanks!
Dan


> Regards,
> Torsten
> 
> 
> On Fri, 2010-01-22 at 14:42 -0800, Dan Williams wrote:
> > On Fri, 2010-01-22 at 11:49 +0100, Torsten Spindler wrote:
> > > Hello,
> > > 
> > > there seems to be a problem for some devices which modemmanager probes
> > > over the serial line. Namely Palm devices [1] and a Smart Card reader
> > > [2]. Is there any possibility to make the probing configurable?
> > > 
> > > There seems to be a work around available by removing 
> > > /usr/lib/ModemManager/libmm-plugin-generic.so. This seems to prevent
> > > modemmanager from doing the autoprobing.
> > 
> > If there are certain devices that are known to handle probing badly,
> > then we can probably blacklist those.  We did that for a few devices
> > with NM 0.7.x but that did not get carried over to ModemManager.  We
> > would however need to get unique identifiers for those devices like USB
> > vid/pid.  Serial devices are harder since legacy serial ports don't
> > really have identifying information.  Be aware that PCMCIA-based cards
> > (like the Sierra 860) will report as ttyS0 using the "serial" driver and
> > thus aren't very distinguishable from legacy serial ports either.  Needs
> > a bit of investigation.
> > 
> > when you get a bug like this, try getting the user to (1) stop NM and
> > MM, (2) plug the device in, and (3) run the "lsudev" utility included in
> > the modem-manager sources in test/ like so:
> > 
> > sudo lsudev tty
> > 
> > and then hit Ctrl+C and get the log into gnome bugzilla.  If we get
> > enough logs we can see if there's a good way to handle legacy serial
> > ports.  If the devices are USB, then that's quite a bit easier.
> > 
> > Dan
> > 
> > 
> 
>