about v3trap behavior in netsnmp - SNMP

This is a discussion on about v3trap behavior in netsnmp - SNMP ; Hi there, I'd noticed that issuing v3 traps in netsnmp will discover engineID but boots/time, and the value 0/0 is carried. Is it correct? BRs...

+ Reply to Thread
Results 1 to 5 of 5

Thread: about v3trap behavior in netsnmp

  1. about v3trap behavior in netsnmp

    Hi there,
    I'd noticed that issuing v3 traps in netsnmp will discover engineID
    but boots/time, and the value 0/0 is carried. Is it correct?

    BRs


  2. Re: about v3trap behavior in netsnmp

    HI,

    How are you generating traps?

    On Tue, 17 May 2005, Zhe-Min Lin wrote:

    > Hi there,
    > I'd noticed that issuing v3 traps in netsnmp will discover engineID
    > but boots/time, and the value 0/0 is carried. Is it correct?
    >
    > BRs


    Regards,
    /david t. perkins

  3. Re: about v3trap behavior in netsnmp

    trapsess -v 3 -u snmpuser -l authPriv -a MD5 -A 12345678 -x DES -X
    12345678 172.23.15.220:162
    => Such specification will result in engineID discovery.

    trapsess -v 3 -e 1234 -u snmpuser -l authPriv -a MD5 -A 12345678 -x DES -X
    12345678 172.23.15.220:162
    => Such specification won't result in engineID discovery but there is
    no boots/time discovery neither.

    Is it correct? I am not sure because it seems reasonable since many
    service provider will setup their own EMS to monitor traps but they
    won't setup a v3 daemon to be discovered. Or is there anything being
    ignored by my fault? My platform is netsnmp 5.2.

    BRs

    dperkins@snmpinfo.com wrote:
    > HI,
    > How are you generating traps?
    > On Tue, 17 May 2005, Zhe-Min Lin wrote:
    > > Hi there,
    > > I'd noticed that issuing v3 traps in netsnmp will discover engineID
    > > but boots/time, and the value 0/0 is carried. Is it correct?
    > > BRs

    > Regards,
    > /david t. perkins


  4. Re: about v3trap behavior in netsnmp


    "Zhe-Min Lin" wrote in message
    news:d6e5p9$2e88$1@netnews2.csie.nctu.edu.tw...

    > Is it correct?


    I think the magic word is "informsession" because net-snmp will translate
    Traps to Informs when using the SNMPv3 protocol. So, you have no Traps, only
    Informs.

    PS: You will need to create an SNMPv3 user at the "snmptrapd end" to recieve
    the Informs - they are authenticated before they are passed to the
    recievers. I do not know the privacy interacts with this.




  5. Re: about v3trap behavior in netsnmp

    Frithiof Andreas Jensen wrote:

    > I think the magic word is "informsession" because net-snmp will translate
    > Traps to Informs when using the SNMPv3 protocol. So, you have no Traps,
    > only Informs.


    That's wrong.

    The Net-SNMP agent will support *both* SNMPv3 Trap and Inform requests,
    depending on whether the "trapsess" entry specified the '-Ci' flag or not.


    > PS: You will need to create an SNMPv3 user at the "snmptrapd end" to
    > recieve the Informs - they are authenticated before they are passed to the
    > recievers.


    That depends on the other options on the "trapsess" line.
    With "-l noAuthNoPriv", then the requests are not authenticated.
    With "-l authNoPriv" or "-l authPriv", they are.

    See
    http://www.net-snmp.org/tutorial/tut...mptrap-v3.html

    for more details.


    > I do not know the privacy interacts with this.


    Privacy is handled in much the same way as authentication.

    Dave


+ Reply to Thread