I ran sysclt -a on the troubled unit and here are the results.  Of particular note are "net.inet.tcp.sack.enable = 1" 
and "net.inet.tcp.congctl.selected = newreno".  If my reading is correct, Fast Retransmission is enabled and the packet 
loss should be detected.

kern.ostype = QNX
kern.osrelease = 6.5.0
kern.version = 2010/07/09-14:44:03EDT
kern.hostname = localhost
kern.clockrate: tick = 8403, hz = 119
kern.mbuf.msize = 256
kern.mbuf.mclbytes = 2048
kern.mbuf.nmbclusters = 8192
kern.mbuf.mblowat = 16
kern.mbuf.mcllowat = 8
kern.sbmax = 262144
net.inet.ip.forwarding = 0
net.inet.ip.redirect = 1
net.inet.ip.ttl = 64
net.inet.ip.forwsrcrt = 1
net.inet.ip.directed-broadcast = 0
net.inet.ip.allowsrcrt = 1
net.inet.ip.subnetsarelocal = 1
net.inet.ip.mtudisc = 1
net.inet.ip.anonportmin = 49152
net.inet.ip.anonportmax = 65535
net.inet.ip.mtudisctimeout = 600
net.inet.ip.maxflows = 256
net.inet.ip.hostzerobroadcast = 1
net.inet.ip.lowportmin = 600
net.inet.ip.lowportmax = 1023
net.inet.ip.maxfragpackets = 200
net.inet.ip.grettl = 0
net.inet.ip.checkinterface = 0
net.inet.ip.ifq.len = 0
net.inet.ip.ifq.maxlen = 256
net.inet.ip.ifq.drops = 0
net.inet.ip.random_id = 0
net.inet.ip.do_loopback_cksum = 0
net.inet.icmp.maskrepl = 0
net.inet.icmp.returndatabytes = 8
net.inet.icmp.errppslimit = 100
net.inet.icmp.rediraccept = 1
net.inet.icmp.redirtimeout = 600
net.inet.icmp.tstamprepl = 1
net.inet.tcp.rfc1323 = 1
net.inet.tcp.sendspace = 22528
net.inet.tcp.recvspace = 32768
net.inet.tcp.mssdflt = 536
net.inet.tcp.syn_cache_limit = 10255
net.inet.tcp.syn_bucket_limit = 105
net.inet.tcp.init_win = 0
net.inet.tcp.mss_ifmtu = 0
net.inet.tcp.sack.enable = 1
net.inet.tcp.sack.maxholes = 32
net.inet.tcp.sack.globalmaxholes = 1024
net.inet.tcp.sack.globalholes = 0
net.inet.tcp.win_scale = 1
net.inet.tcp.timestamps = 1
net.inet.tcp.compat_42 = 0
net.inet.tcp.cwm = 0
net.inet.tcp.cwm_burstsize = 4
net.inet.tcp.ack_on_push = 0
net.inet.tcp.keepidle = 14400
net.inet.tcp.keepintvl = 1
net.inet.tcp.keepcnt = 20
net.inet.tcp.slowhz = 2
net.inet.tcp.log_refused = 0
net.inet.tcp.init_win_local = 4
net.inet.tcp.rstppslimit = 100
net.inet.tcp.delack_ticks = 23
net.inet.tcp.hiwat_adjust = 1
net.inet.tcp.do_loopback_cksum = 0
net.inet.tcp.congctl.available = reno newreno
net.inet.tcp.congctl.selected = newreno
net.inet.tcp.ecn.enable = 0
net.inet.tcp.ecn.maxretries = 1
net.inet.tcp.iss_hash = 0
net.inet.tcp.abc.enable = 1
net.inet.tcp.abc.aggressive = 1
net.inet.udp.checksum = 1
net.inet.udp.sendspace = 9216
net.inet.udp.recvspace = 41600
net.inet.udp.do_loopback_cksum = 0
net.inet.arp.prune = 300
net.inet.arp.keep = 1118
net.inet.arp.down = 20
net.inet.arp.refresh = 300
net.bpf.maxbufsize = 1048576
hw.machine = x86pc
hw.machine_arch = x86
hw.alignbytes = 3
qnx.kern.enmap = 1
qnx.net.inet.tcp.hiwat_adjust = 1

Fast retransmission is supported. It is triggered by a 3 duplicate ACKs being received. If your application isn't 
sending enough data that there are 3 more segments in flight after the lost segment to trigger the 3 duplicate ACKs then
 that would be why you are not seeing the fast retransmission.

Do you have a wireshark trace showing the 3 duplicate ACKs?