On systemd < v248, there is a bug with linux 5.11 and higher that causes systemd-rfkill service to fail to start. The fix is pulled into v248 - can =sys-apps/systemd-248 be stabilized to pull in the fix? See https://github.com/systemd/systemd/issues/18677 and https://github.com/systemd/systemd/pull/18679 Reproducible: Always Steps to Reproduce: 1. Install gentoo with systemd, use =sys-apps/systemd-247.2-r4 (current stable version) 2. Install kernel version 5.11 or higher 3. Boot up Actual Results: $ systemctl --failed UNIT LOAD ACTIVE SUB DESCRIPTION ● systemd-rfkill.service loaded failed failed Load/Save RF Kill Switch Status LOAD = Reflects whether the unit definition was properly loaded. ACTIVE = The high-level unit activation state, i.e. generalization of SUB. SUB = The low-level unit activation state, values depend on unit type. 1 loaded units listed. Expected Results: $ systemctl --failed UNIT LOAD ACTIVE SUB DESCRIPTION 0 loaded units listed. $ uname -r 5.12.2-gentoo $ emerge -pv1 systemd These are the packages that would be merged, in order: Calculating dependencies... done! [ebuild R ] sys-apps/systemd-247.2-r4:0/2::gentoo USE="acl cryptsetup gcrypt hwdb kmod lz4 pam pcre policykit resolvconf seccomp (split-usr) sysv-utils zstd -apparmor -audit -build -cgroup-hybrid -curl -dns-over-tls -elfutils -gnuefi -homed -http -idn -importd -lzma -nat -pkcs11 -pwquality -qrcode -repart (-selinux) -static-libs -test -vanilla -xkb" ABI_X86="32 (64) (-x32)" 0 KiB Total: 1 package (1 reinstall), Size of downloads: 0 KiB
Note that 5.11 is not stable in Gentoo but we can’t really control which kernels users are on, so I guess this is worth thinking about.
Per https://github.com/systemd/systemd/issues/18677 this bug also affects linux 5.10 (though the service does not fail to start) The following can be seen in the journal: systemd-rfkill[pid]: Couldn't write rfkill event structure, too short.
