login: alc_callout_replica: loaded (hz=100 ncpus=6)
alcreplica: armed (hz=100)
alcreplica[try]: stop (fired=2 ran=2)
alcreplica[try]: callout_stop returned, pending=0
alcreplica[try]: RESULT pending=0 (0 == try-shape fully stopped, no deadlock)
alcreplica[blk]: alc_tick handler BLOCKED on our serializer (fired=13 ran=12) - entering callout_stop()
alcreplica[blk]: IF THIS IS THE LAST alcreplica LINE, THE if_alc.c AB-BA DEADLOCK IS PROVEN
% Wedged-thread evidence (second session, fresh ssh):
%   $ ps -o pid,stat,wchan,command -ax | grep stop_blk
  1334 D5     costp    sysctl kern.alcreplica.stop_blk=1
%
% ^ D-state (uninterruptible) wchan "costp" == ssleep label inside
% _callout_cancel_or_stop() sys/kern/kern_timeout.c:923 - the alc_stop()-
% shaped caller is permanently stuck inside callout_stop() HOLDING the
% serializer; the alc_tick()-shaped callback is parked in
% lwkt_serialize_enter (if_alc.c:3414 protocol) waiting for that
% serializer.  AB-BA complete.  First ssh session attempting follow-up
% commands also hung (collateral); a fresh session worked for ps.