Fedora 33 brough systemd-resolved by default; but in Fedora 35 this
somehow got reverted.
I've proposed it as a blocker, but the main point of the thread is
really to discuss the general case of whether such a thing is a
blocker? I'm not thinking of a release criterion that applies to this
case. It seems reasonable that approved+implemented features that
subsequently break (accidentally or even intentionally when absent an
approved change) should be blockers.
Server edition is missing /etc/resolv.conf symlink (use systemd-resolved)
https://bugzilla.redhat.com/show_bug.cgi?id=2032085
--
Chris Murphy