Description of problem:

After an update to 10-2.fc (that includes systemd-units and systemd-sysvinit) networking is not brought up anymore.  There is a looong timeout on a failed attempt to bring back a loopback interface and it is downhill from there.  Another very long timeout, not recorded in a boot.log, occures with 
readahead-collect.service.  These messages show up but do not end up in boot.log:

init[1]: Unit readahead-collect.service entered failed state.
init[1]: rtkit-daemon.service operation timed out. Terminating.
init[1]: network.service operation timed out. 
init[1]: Unit network.service entered failed state.
init[1]: restorecond.service: control process exited, code=exited status=7
init[1]: Unit restorecond.service entered failed state.

Version-Release number of selected component (if applicable):
systemd-10-2.fc15.x86_64

How reproducible:
always

Additional info:
Booting with 'init=/sbin/upstart' makes a machine boot correctly (without downgrading any packages).  The other way work around the issue is to downgrade systemd packages to 9-3.fc15.

Although these

init[1]: Unit readahead-collect.service entered failed state.
init[1]: restorecond.service: control process exited, code=exited status=7
init[1]: Unit restorecond.service entered failed state.

show up with systemd-9-3.fc15 too there is no very long timeout on readahead-collect.service.  Besides I thought that "status=7" was suppsed to be "special" and this should be silent.  On my test system selinux is off so this particular failure is fully expected.