[SNMP4J] Strange problem with multithreaded SNMPv3-Scan

[chk-world at gmx.de]

Hi SNMP4J Experts,

I'm using SNMP4J (v. 1.0.3) and try to programm a multithreaded SNMP (v3)
scanner. It should scan a network for instance between 192.168.1.1 and
192.168.1.255. That means sending a v3PDU and waiting for the answer or a
timeout. 
Because of the long time it would take to send all the packets in a single
thread via snmp.sendPDU() I'm using a thread pool.

In that network I have a correctly configured snmp v3 agent (IP
192.168.1.5). Asking only this host (with v3 pakets) works well.

Using a timeout of 1000 ms I get the following results from a the above host
when scanning from IP 1 to 20.

a: Request timed out.
b: 1.3.6.1.2.1.1.5.0 = E1_2 (my sysName)
c: REPORT: Message not within time window. Current counter value is 359
d: REPORT: Wrong digest. Current counter value is 23.

Setting the timeout up to 3000 ms doesn't help (than only timeouts come).
Ethereal shows the following for case b:

#1 manager -> agent : GET
#2 agent -> manager : usmStatsUnknownEngineIDs
#3 manager -> agent : Encrypted PDU
#4 agent -> manager : usmStatsNotInTimeWindow
#5 manager -> agent : Encrypted PDU
#6 agent -> manager : Encrypted PDU (I suppose the right answer)

In case "a" only #1 to #4 are displayed by ethereal.

Does anyone have an idea what could be wrong?
Thanks in advance,

Chris



-- 
NEU +++ DSL Komplett von GMX +++ http://www.gmx.net/de/go/dsl
GMX DSL-Netzanschluss + Tarif zum supergünstigen Komplett-Preis!