[SNMP4J] Question about receiving SNMPv3 traps

[Girish Venkatasubramanian]

Hello
I am seeking some clarification about receiving SNMPv3 traps using SNMP4J.

In my setup, I have a few network devices configured for SNMPv3. They all
use the *same* userName and auth and priv keys. They are configured to use
authentication and privacy when sending traps.

In my trap receiver, which I plan to write using SNMP4J, the userName and
credentials are known.

1) In order to receive the traps from these devices, is it sufficient to
add an entry in the USM cache of my TrapReceiver as below  ?

*snmp*.getUSM().addUser( *new *OctetString(username),

                       *new *UsmUser(*new *OctetString(username), AuthMD5.
*ID*, *new *OctetString(authpassphrase), PrivAES128.*ID*, *new *
OctetString(privacypassphrase)));


>From this email thread,
http://oosnmp.net/pipermail/snmp4j/2013-April/005042.html I quote
"

If you set the autoDiscovery property of the USM to true, then it is
even easier.
You do not have to add localized USM users (thus you do not have to know the
engineIDs of the notification senders), you simply add the users without
engineID.

"
To clarify, do I need to set the autoDiscovery property on the USM of the
trap receiver ?

2) If I have to add a localized user for each of these trap senders, then
should Snmp.discoverAuthoritativeEngineID(..) be used and then a
per-trap-sender target be added using this engine ID ? That is, the key
localization approach as shown
https://doc.snmp.app/pages/viewpage.action?pageId=1441800

Ideally, I would like to avoid having to discover the authoritative engine
ID as that will restrict the ability of my trap receiver to receive traps
only from known targets . Would having a different userName (but same
authentication and privacy keys) for each of the trap sender help ?

Any clarifications on these would be greatly appreciated.
Thanks
Girish