> I'm just wondering, isn't it too late for this in anaconda? We > depend on udev to detect the devices in loader (see > hardware.c:detectHardware) and before systemd we set this in init.c. > The question is if we need those rules so early or not. As long as it comes before the udev_trigger call in filter_gui.py, it shouldn't be too late. Of course, earlier would be better. > If this is about some kind of systemd concurrency business then I > think systemd should provide something that will let us to wait for > the udev daemeon to be up before we call detectHardware() for the > first time. The problem is entirely caused by me moving the udev startup from init.c to just letting the service start up. We were adding --env=ANACONDA=1 to our udev arguments. Now there's nowhere to add udev arguments. Now that I think about it, I could add this way early in loader.c instead. I guess my instincts just told me to not put anything in loader. - Chris _______________________________________________ Anaconda-devel-list mailing list Anaconda-devel-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/anaconda-devel-list
