Approach to system testing for Osmocom stack

This is merely a historical archive of years 2008-2021, before the migration to mailman3.

A maintained and still updated list archive can be found at

Holger Freyther holger at
Mon Aug 20 08:47:58 UTC 2018

> On 10. Jan 2018, at 09:49, Pau Espin Pedrol <pespin at> wrote:

Hey Pau,

> I never used virtbts but I think it should be quite easy adding support for it in osmo-gsm-tester. The only big issue I can think of is that we need to find a way to differentiate between virtbs and other bts since they use a different communication medium and then that means that the requested modems to run for the test with virtbts need to be selected accordingly (ie. don't allocate an ofono modem from the object pool if we are using a virbts). That can probably accomplished easily too, by adding new attributes to objects to describe that characteristic.
> It can be a good idea to support this at some point if we want to test for compatibility between "mobile" app + virtbs and other BTS and modems, for instance see if we can place a call between a regular modem with a sysmobts and a "mobile" app + virtbts. Then we can make sure we don't add regressions in the future.

I think I have finally a routine that gives me some time to move this forward. We should have the ability for bi-directional communication between mobile and the test driver and can start implementing the mobile as a non-ofono "modem".

From a mobile lua/point of view: The mobile will start, register over unix dgram to a server. At this point the server can send commands (as JSON) and the lua script can dispatch them (e.g. on/off, cell selection, send sms, place call)

Can you help me to see what needs to be done from a osmo-gsm-tester point of view? What amount of work is necessary? The benefit might be to have a quick(er) test on Jenkins.


More information about the baseband-devel mailing list