During testing of the cluster demo setup with tbtnavi, it was noticed that cluster-receiver on the cluster board always crashes when the IVI board is rebooted.  The crash seems to happen when the RTP stream starts again after the IVI board has booted.  It seems likely that the switch to RTP streaming with the Weston remoting feature instead of the previous H.264 with gst-record is responsible.  Logs:

Oct 19 21:47:32 raspberrypi3 afbd-cluster-receiver@1.0-60b6e19[269]: The surface with id: 2000000 does not exist
Oct 19 21:47:32 raspberrypi3 afbd-cluster-receiver@1.0-60b6e19[269]: surface_get: the surface with given id does not exist
Oct 19 21:47:32 raspberrypi3 kernel: audit: type=1701 audit(1571521652.109:5): auid=4294967295 uid=1001 gid=1001 ses=4294967295 subj=User::App::cluster-receiver pid=276 comm="cluster-receive" exe="/var/local/lib
/afm/applications/cluster-receiver/1.0-60b6e19/bin/cluster-receiver" sig=11 res=1
Oct 19 21:47:32 raspberrypi3 audit[276]: ANOM_ABEND auid=4294967295 uid=1001 gid=1001 ses=4294967295 subj=User::App::cluster-receiver pid=276 comm="cluster-receive" exe="/var/local/lib/afm/applications/cluster-r
eceiver/1.0-60b6e19/bin/cluster-receiver" sig=11 res=1
Oct 19 21:47:32 raspberrypi3 afbd-cluster-receiver@1.0-60b6e19[269]: free(): invalid next size (fast)
Oct 19 21:47:32 raspberrypi3 systemd[1]: Created slice system-systemd\x2dcoredump.slice.
Oct 19 21:47:32 raspberrypi3 systemd[1]: Started Process Core Dump (PID 371/UID 0).
Oct 19 21:48:49 raspberrypi3 systemd-coredump[372]: Process 276 (cluster-receive) of user 1001 dumped core.