Skip to content

bridge: test connection.autoconnect-slaves behavior

Thomas Haller requested to merge th/autoconnect-slaves-rh1548265 into master

autoconnect.slaves should forcefully activate all slaves, even if the device is currently busy with another profile. That was the case already, but slightly broken with [1].

This adds two tests, one a general test, and the other where the profile that should be stolen for activation is the slave profile itself, but activated previously as a non-slave before being modified. This is the scenario of bug [2].

[1] https://cgit.freedesktop.org/NetworkManager/NetworkManager/commit/?id=0922a177385be188b9c9c8ad39c1068533f5a4b3 [2] https://bugzilla.redhat.com/show_bug.cgi?id=1548265

The second test is not yet expected to pass on master, which is what [2] is all about.

Merge request reports