Enterprise Manager

I'm starting a new thread on this ... this was begun the thread about recvlets.

I'm on Linux (Redhat ES4) and I am working on extending OEM to deal with a lot of SNMP devices. I need to be able to receive traps; and for this I have defined a recvlet "SNMPTrap". However, I am not able to get the snmp software (master + encap) to listen to port udp/162 for trap messages? It listens fine on 161 but not 162?

CONFIG.encap:

AGENT AT PORT 1161 WITH COMMUNITY apctrap
SUBTREES     1.3.6.1.4.1.318
FORWARD ALL TRAPS;

And the 1161 port is reflected in start_peer as specified.

CONFIG.master:

COMMUNITY       apctrap
                ALLOW ALL OPERATIONS
                USE NO ENCRYPTION
                MEMBERS localhost, 172.16.52.60

MANAGER         localhost
                SEND NO TRAPS

MANAGER         172.16.52.60
                SEND NO TRAPS

Finally, /etc/snmp/snmpd.conf (not the snmpd.conf that's in $AGENT_HOME/network/snmp/peer):

com2sec apcTrap        default       apctrap
group   apc            v1            apcTrap
view    systemview    included   .1.3.6.1.4.1.318
access  apcTrap        ""      any       noauth    exact  systemview none none

I run start_peer -a as root, and the emctl start subagent, as "oracle" - and all processes start up and remain running.

Netstat reports:

tcp        0      0 0.0.0.0:199                 0.0.0.0:*                   LISTEN      13323/master_peer
udp        0      0 172.16.52.80:1160           0.0.0.0:*                               13327/encap_peer
udp        0      0 0.0.0.0:1162                0.0.0.0:*                               13327/encap_peer
udp        0      0 0.0.0.0:161                 0.0.0.0:*                               13323/master_peer
udp        0      0 0.0.0.0:1161                0.0.0.0:*                               13335/snmpd

(I have manually filtered all the non snmp ports out).
As the list shows, only 161 is being used - not 162 which is the trap receiption port?

Am I missing something?