[systemd-devel] udev database backwards compatibility guarantees

[Kay Sievers]

On Mon, Oct 27, 2014 at 4:25 PM, Lennart Poettering
<lennart at poettering.net> wrote:
> On Mon, 27.10.14 14:43, Kay Sievers (kay at vrfy.org) wrote:
>
>> On Mon, Oct 27, 2014 at 2:22 PM, Alexander Larsson <alexl at redhat.com> wrote:
>> > On lör, 2014-10-25 at 13:45 +0200, Kay Sievers wrote:
>> >> > Kay, any ideas on the udev database stability?
>> >>
>> >> No stability. And so far no guarantees that things will not change.
>> >>
>> >> The versions of the udev daemon, libudev and the runtime data must
>> >> match. Any expectations about version mix and match would require a
>> >> promise we do not give at this moment.
>> >>
>> >> It might change with an imaginary "sd-device library", but it is very
>> >> unlikely to happen with the current udev.
>> >
>> > So, libudev will not be supportable as bundled in a sandboxed app then?
>>
>> Right. I don't think we can make any such promises with the current code base.
>>
>> The event monitor depends on a the same version of the running daemon,
>> or the properties which depend on the data in /run. Only the part that
>> finds devices, reads properties and enumerates /sys should be fine.
>
> Maybe the right approach is to move things over to kdbus, and then
> make sandboxed apps use the kdbus interface, and hide the other stuff
> away or so...

Yeah, that sounds like the way to go. We should be very careful with
exposing the so-far-private /run state to the sandboxed apps if we
want to make promises about backwards compat. I don't think that will
work out in the longer run.

Kay