Even trivial service occasionally get stuck, for example when
there's a problem with the journal. There's nothing more annoying
that looking at the cylon eye for a job with an infinite timeout.
Use standard 90s for jobs that do some work, and 30s for those which
should be almost instantenous.
Type=oneshot
RemainAfterExit=yes
ExecStart=@rootlibexecdir@/systemd-binfmt
+TimeoutSec=90s
Type=oneshot
RemainAfterExit=yes
ExecStart=@rootbindir@/systemd-hwdb update
+TimeoutSec=90s
Type=oneshot
RemainAfterExit=yes
ExecStart=@rootlibexecdir@/systemd-machine-id-commit
+TimeoutSec=30s
Type=oneshot
RemainAfterExit=yes
ExecStart=@rootlibexecdir@/systemd-modules-load
+TimeoutSec=90s
RemainAfterExit=yes
ExecStart=@rootlibexecdir@/systemd-random-seed load
ExecStop=@rootlibexecdir@/systemd-random-seed save
+TimeoutSec=30s
RemainAfterExit=yes
ExecStart=@rootlibexecdir@/systemd-rfkill load %I
ExecStop=@rootlibexecdir@/systemd-rfkill save %I
+TimeoutSec=30s
Type=oneshot
RemainAfterExit=yes
ExecStart=@rootlibexecdir@/systemd-sysctl
+TimeoutSec=90s
Type=oneshot
RemainAfterExit=yes
ExecStart=@rootbindir@/systemd-sysusers
+TimeoutSec=90s