The update from sys-fs/udev-238 to sys-fs/udev-239 breaks network on my systems, because they're renamed from enp0s3 to ens3 on my virtual machines (libvirt/kvm). Downgrading to sys-fs/udev-238 fixed this reliable.
Please test with udev-240. If this is still an issue, please report it upstream [1] and link the issue here. [1] https://github.com/systemd/systemd/issues/
With udev-240 it's renamed again, to eth0. So now we have: udev-238: enp0s3 udev-239: ens3 udev-240: eth0 This was all tested with linux-4.19.12. The only part that was changed was udev and meson for build-reasons.
bug reported upstream: https://github.com/systemd/systemd/issues/11277
According to upstream, the network interface name from udev-238 was buggy. So changing this is right. But since udev-238 shows the wrong behaviour a big warning should be displayed during update. Especially for headless systems it's not acceptable, when the network interface becomes unreachable because of an update.
The behavior in 239 is correct, and 240 has also been fixed via a backported patch.