Making disconnection more robust

[Aleksander Morgado]

On Fri, Aug 11, 2017 at 3:58 PM, Tim Small <tim at seoss.co.uk> wrote:
> On 11/08/17 14:09, Aleksander Morgado wrote:
>> I'd suggest we try to move
>> that modem to "Failed" state instead, as we really lost the control
>> port; and let the applications using MM do whatever they need to do
>> with a failed modem, including external power cycle or whatever.
>
> Hi,
>
> Yes that makes more sense.  That had crossed my mind previously, but I
> didn't look into it further.
>
> Presumably it'd be possible to create a separate service which looks for
> a particular modem failing, and takes the necessary action to reset and
> trigger a reprobe etc.?  Would there be any issue with this running at
> the same time as nm for instance?
>

No issues, no. NM should be ok if a modem suddenly goes into Failed
state and then restarted by some other service. If you want to script
the other service, just a bash script running mmcli calls or some
python program (e.g. using GObject Introspection) would be very easy
to do, see examples/ directory for some examples.

-- 
Aleksander
https://aleksander.es